JP7541550B2 - Targeted Disruption of MHC Cell Receptors - Google Patents

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 62/269,410, filed December 18, 2015, U.S. Provisional Application No. 62/305,097, filed March 8, 2016, and U.S. Provisional Application No. 62/329,439, filed April 29, 2016, the disclosures of which are incorporated herein by reference in their entireties.

TECHNICAL FIELD This disclosure is in the field of genomic modification of human cells, including lymphocytes and stem cells.

2. Background Gene therapy holds enormous potential for a new era of human therapeutics. These methodologies allow treatment for conditions that have not been addressable by standard medical practice. Gene therapy can include many variations of genome editing techniques, such as disruption or correction of gene loci, and insertion of expressible transgenes that can be regulated by specific exogenous promoters fused to the transgene, or by endogenous promoters found at the site of insertion into the genome.

Transgene delivery and insertion are examples of obstacles that must be overcome for any true implementation of this technology. For example, a variety of gene delivery methods are potentially available for therapeutic use, but all involve substantial trade-offs between safety, durability, and expression levels. Methods that provide transgenes as episomes (e.g., basic adenovirus (Ad), adeno-associated virus (AAV), and plasmid-based systems) are generally safe and can yield high initial expression levels, but these methods lack robust episomal replication, which may limit the duration of expression in mitotically active tissues. In contrast, delivery methods that result in random integration of the desired transgene (e.g., lentivirus (LV) integration) provide more persistent expression, but the non-targeted nature of random insertion can cause unregulated growth in recipient cells and may result in malignant disease through activation of oncogenes in the vicinity of the randomly integrated transgene cassette. Furthermore, even if transgene integration avoids replication-driven loss, it does not prevent eventual silencing of the exogenous promoter fused to the transgene. Over time, such silencing results in reduced transgene expression in the majority of non-specific insertion events. Moreover, transgene integration rarely occurs in all target cells, which can make it difficult to achieve sufficiently high expression levels of the transgene of interest to achieve the desired therapeutic effect.

Recently, new strategies for transgene integration have been developed that use cleavage by site-specific nucleases (e.g., zinc finger nucleases (ZFNs), transcription activator-like effector domain nucleases (TALENs), CRISPR/Cas and Cfp1/CRISPR systems with engineered crRNA/tracr RNA ("single guide RNA") to guide specific cleavage) to bias insertion into a selected genomic locus. For example, U.S. Patent Nos. 9,255,250; 9,045,763; 9,005,973; 8,956,828; 8,945,868; 8,703,489; 8,586,526; 6,534,261; 6,599,692; 6,503,717; 6,689,558; 7,067,317; 7,262,054; 7,888,121; 7,972,854; 7,914,796; 7,951,925; 8,110,379; 8,40 See U.S. Patent Application Publication Nos. 20030232410; 20050208489; 20050026157; 20050064474; 20060063231; 20080159996; 201000218264; 20120017290; 20110265198; 20130137104; 20130122591; 20130177983; and 20130177960; and 20150056705. Additionally, targeted nucleases have been developed based on the Argonaute system (e.g., from T. thermophilus, known as "TtAgo," see Swarts et al. (2014) Nature 507(7491):258-261), which may also have potential for use in genome editing and gene therapy. Compared to classical integration approaches, this nuclease-mediated approach to transgene integration offers the promise of improved transgene expression, increased safety, and persistence of expression, since it allows for precise transgene placement with minimal risk of gene silencing or activation of nearby oncogenes.

The T cell receptor (TCR) is an essential part of the selective activation of T cells. With some similarity to antibodies, the antigen recognition portion of the TCR is typically made of two chains, α and β, which coassemble to form a heterodimer. The antibody similarity lies in the manner in which a single gene is assembled that codes for the TCR alpha and beta complex. The TCR alpha (TCRα) and beta (TCRβ) chains are each composed of two regions, a C-terminal constant region and an N-terminal variable region. The genomic loci that code for the TCR alpha and beta chains are similar to those that code for antibodies in that the TCRα gene contains V and J segments, while the β chain locus contains a D segment in addition to the V and J segments. For the TCRβ locus, there are additionally two different constant regions to be selected from during the selection process. During T cell development, the various segments recombine such that each T cell contains a unique TCR variable portion, called the complementarity determining region (CDR), in the alpha and beta chains, and the body has a large repertoire of T cells that are capable of interacting with a unique antigen presented by an antigen presenting cell because of their unique CDR. When TCRα or β gene rearrangement occurs, expression of the second corresponding TCRα or TCRβ is suppressed so that each T cell expresses only one unique TCR structure in a process called "antigen receptor allelic exclusion" (see Brady et al., (2010) J Immunol 185:3801-3808).

During T cell activation, the TCR interacts with antigens that are presented as peptides on the major histocompatibility complex (MHC) of antigen-presenting cells. Recognition of the antigen-MHC complex by the TCR results in T cell stimulation, which in turn leads to differentiation into both memory and effector lymphocytes, helper T cells (CD4+) and cytotoxic T lymphocytes (CD8+). These cells can then be expanded in a clonal manner to provide activated subpopulations within the total T cell population that are capable of responding to certain antigens.

There are two classes of MHC proteins, I and II. Class I MHC proteins are heterodimers of two proteins, the α chain, a transmembrane protein encoded by the MHC1 class I gene, and the β2 microglobulin chain (sometimes called B2M), a small extracellular protein encoded by a gene not present in the MHC gene cluster. The α chain folds into three globular domains, and when the β2 microglobulin chain associates, the globular structural complex resembles an antibody complex. The foreign peptide is present on the two most N-terminal domains, which are also the most variable. Class II MHC proteins are also heterodimers, but the heterodimer contains two transmembrane proteins encoded by genes within the MHC complex. Class I MHC:antigen complexes interact with cytotoxic T cells, while class II MHC presents antigen to helper T cells. Furthermore, while class I MHC proteins tend to be expressed in almost all nucleated cells and platelets (and in mice, red blood cells), class II MHC proteins are more selectively expressed. Typically, class II MHC proteins are expressed in B cells, some macrophages and monocytes, Langerhans cells, and dendritic cells.

In humans, the class I HLA gene cluster contains three major loci, B, C and A, and several minor loci, including E, G and F, all found in the HLA region on chromosome 6. The class II HLA cluster also contains three major loci, DP, DQ and DR, and both the class I and class II gene clusters are polymorphic in that there are several different alleles of both class I and II genes within the population. There are also several accessory proteins that play a role in HLA function as well. Beta-2 microglobulin functions as a chaperone (encoded by B2M and located on chromosome 15) and stabilizes HLA A, B or C proteins expressed on the cell surface, and also stabilizes antigens that display grooves on the class I structure. It is normally found in small amounts in serum and urine.

HLA plays a major role in graft rejection. The acute phase of transplant rejection may occur within about 1-3 weeks and usually involves the action of host T lymphocytes on the donor tissue due to sensitization of the host system to donor class I and class II HLA molecules. In many cases, the inciting antigen is class I HLA. For best success, the donor is typed for HLA and matched as perfectly as possible to the patient-recipient. However, even donation between family members that may share a high percentage of HLA identity is often unsuccessful. Therefore, to keep the transplanted tissue in the recipient, the patient must often be subjected to intensive immunosuppressive therapy to prevent rejection. Such treatment may result in complications and significant morbidity due to opportunistic infections that the patient may have difficulty overcoming. Regulation of class I or II genes may be disrupted in the presence of some tumors, and such disruption may have causal consequences for the patient's prognosis. For example, reduced B2M expression was found in metastatic colorectal cancer (Shrout et al. (2008) Br J Canc 98:1999). Because B2M has a key role in stabilizing the MHC class I complex, it was hypothesized that reduced B2M in certain solid cancers is a mechanism of immune escape from T cell-driven immune surveillance. Reduced B2M expression has been shown to be the result of specific mutations in the B2M coding sequence that result in suppression and/or genetic knockout of normal IFN-gamma B2M expression control (Shrout et al., ibid.). Confusingly, increased B2M has also been associated with some types of cancer. Increased urinary B2M levels serve as a prognostic factor for several cancers, including prostate, chronic lymphocytic leukemia (CLL) and non-Hodgkin's lymphoma.

Adoptive cell therapy (ACT) is an evolving form of cancer treatment based on delivering tumor-specific immune cells to a patient so that the delivered cells attack and eliminate the patient's cancer. ACT may involve the use of tumor-infiltrating lymphocytes (TILs), which are T cells isolated from the patient's own tumor mass and expanded ex vivo for reinfusion back into the patient. This approach has shown promise in treating metastatic melanoma, with one study observing a long-term response rate of >50% (see, e.g., Rosenberg et al. (2011) Clin Canc Res 17(13):4550). TILs are a promising source of cells because they are a mixed set of the patient's own cells that bear T cell receptors (TCRs) specific for tumor-associated antigens (TAAs) present on the tumor (Wu et al. (2012) Cancer J 18(2):160). Other approaches involve editing T cells isolated from a patient's blood so that they are engineered to be responsive in some way to the tumor (Kalos et al., (2011) Sci Transl Med 3(95):95ra73).

Chimeric antigen receptors (CARs) are molecules designed to target immune cells to specific molecular targets expressed on the cell surface. In their most basic form, they are receptors introduced into cells that couple a specificity domain expressed on the outside of the cell to a signaling pathway inside the cell so that the cell becomes activated when the specificity domain interacts with its target. Often CARs are made from mimicking functional domains of T cell receptors (TCRs) in which an antigen-specific domain, such as an scFv or some type of receptor, is fused to a signaling domain, such as an ITAM and other costimulatory domains. These constructs are then introduced into T cells ex vivo, allowing the T cells to become activated in the presence of cells expressing the target antigen, resulting in the attack of target cells by the activated T cells in a non-MHC-dependent manner when the T cells are reintroduced into the patient (see Chicaybam et al. (2011) Int Rev Immunol 30:294-311; Kalos, ibid.). Thereby, adoptive cell therapy using T cells modified ex vivo with engineered TCRs or CARs is a very promising clinical approach for several types of diseases. For example, targeted cancers and their antigens include follicular lymphoma (CD20 or GD2), neuroblastoma (CD171), non-Hodgkin's lymphoma (CD19 and CD20), lymphoma (CD19), glioblastoma (IL13Rα2), chronic lymphocytic leukemia or CLL and acute lymphocytic leukemia or ALL (both CD19). Virus-specific CARs have also been developed to attack cells carrying viruses such as HIV. For example, clinical trials have been initiated using CARs specific for Gp100 for the treatment of HIV (Chicaybam, ibid.).

ACTR (antibody-binding T cell receptor) is an engineered T cell component that is capable of binding to exogenously supplied antibodies. Binding of the antibody to the ACTR component prepares the T cell to interact with the antigen recognized by the antibody, and upon antigen encounter, T cells containing ACTR become capable of interacting with the antigen (see US Patent Publication No. 20150139943).
However, one of the drawbacks of adoptive cell therapy is that the source of the cell product must be patient-specific (autologous) to avoid potential rejection of the transplanted cells. This has led researchers to develop methods to edit the patient's own T cells to avoid this rejection. For example, the patient's T cells or hematopoietic stem cells can be manipulated ex vivo by the addition of engineered CARs, ACTRs and/or T cell receptors (TCRs), and then further treated with engineered nucleases to knock out T cell checkpoint inhibitors such as PD1 and/or CTLA4 (see PCT Publication WO2014/059173). For application of this technology to a larger patient population, it is advantageous to develop a universal population of cells (allogeneic). Furthermore, knocking out the TCR results in cells that cannot mount a graft-versus-host disease (GVHD) response when introduced into the patient.
Thus, there remains a need for methods and compositions that can be used to modify MHC gene expression (e.g., knock out B2M) and/or knock out TCR expression in T cells.

U.S. Pat. No. 9,255,250 U.S. Pat. No. 9,045,763 U.S. Pat. No. 9,005,973 U.S. Pat. No. 8,956,828 U.S. Pat. No. 8,945,868 U.S. Pat. No. 8,703,489 U.S. Pat. No. 8,586,526 U.S. Pat. No. 6,534,261 U.S. Pat. No. 6,599,692 U.S. Pat. No. 6,503,717 U.S. Pat. No. 6,689,558 U.S. Pat. No. 7,067,317 U.S. Pat. No. 7,262,054 U.S. Pat. No. 7,888,121 U.S. Pat. No. 7,972,854 U.S. Pat. No. 7,914,796 U.S. Pat. No. 7,951,925 U.S. Pat. No. 8,110,379 U.S. Pat. No. 8,409,861 US Patent Application Publication No. 2003/0232410 US Patent Application Publication No. 2005/0208489 US Patent Application Publication No. 2005/0026157 US Patent Application Publication No. 2005/0064474 US Patent Application Publication No. 2006/0063231 US Patent Application Publication No. 2008/0159996 US Patent Application Publication No. 2010/00218264 US Patent Application Publication No. 2012/0017290 US Patent Application Publication No. 2011/0265198 US Patent Application Publication No. 2013/0137104 US Patent Application Publication No. 2013/0122591 US Patent Application Publication No. 2013/0177983 US Patent Application Publication No. 2013/0177960 US Patent Application Publication No. 2015/0056705 US Patent Application Publication No. 2015/0139943

Swarts et al. (2014) Nature 507(7491):258-261 Brady et al. (2010) J Immunol 185:3801-3808 Shrout et al. (2008) Br J Canc 98:1999 Rosenberg et al. (2011) Clin Canc Res 17(13):4550 Wu et al. (2012) Cancer J 18(2):160 Kalos et al. (2011) Sci Transl Med 3(95):95ra73 Chicaybam et al. (2011) Int Rev Immunol 30:294-311

SUMMARY Disclosed herein are compositions and methods for partial or complete inactivation or disruption of the B2M gene, as well as compositions and methods for the introduction and desired level expression of exogenous TCR, CAR or ACTR transgenes in T lymphocytes following or concomitant with disruption of endogenous TCR and/or B2M. Additionally provided herein are methods and compositions for deleting (inactivating) or suppressing the B2M gene to produce HLA class I null T cells, stem cells, tissues or whole organisms, e.g., cells that do not express one or more HLA receptors on their surface. In certain embodiments, the HLA null cells or tissues are human cells or tissues that are advantageous for use in transplantation. In a preferred embodiment, the HLA null T cells are prepared for use in adoptive T cell therapy.

In one aspect, described herein is an isolated cell (e.g., a eukaryotic cell, such as a mammalian cell, including a lymphoid cell, a stem cell (e.g., an iPSC, an embryonic stem cell, an MSC, or an HSC), or a progenitor cell) in which expression of the beta 2 microglobulin (B2M) gene is modulated by a modification of exon 1 or exon 2 of the B2M gene. In certain embodiments, the modification is within the sequence set forth in one or more of SEQ ID NOs: 6-48 and/or 137-205; within 1-5 base pairs, 1-10 base pairs, or 1-20 base pairs on either side (flanking genomic sequence) of SEQ ID NOs: 6-48 or 137-205; or within GGCCTTA, TCAAAT, TCAAATT, TTACTGA, and/or AATTGAA. The modification may be by an exogenous fusion molecule comprising a functional domain (e.g., a transcriptional regulatory domain, a nuclease domain) and a DNA binding domain, including, but not limited to: (i) a cell comprising an exogenous transcription factor comprising a DNA binding domain that binds to a target site set forth in any of SEQ ID NOs: 6-48 and/or 137-205 and a transcriptional regulatory domain, whereby the transcription factor modifies B2M gene expression, and/or (ii) a cell comprising an insertion and/or deletion within one or more of SEQ ID NOs: 6-48 and/or 137-205; within 1-5 base pairs, 1-10 base pairs, or 1-20 base pairs on either side (flanking genomic sequences) of SEQ ID NOs: 6-48 and/or 137-205; or within GGCCTTA, TCAAAT, TCAAATT, TTACTGA, and/or AATTGAA. The cells may include further modifications, such as an inactivated T cell receptor gene, a PD1 and/or CTLA4 gene and/or a transgene encoding a chimeric antigen receptor (CAR), a transgene encoding an antibody-binding T cell receptor (ACTR) and/or a transgene encoding an antibody. Pharmaceutical compositions comprising any of the cells described herein and methods of using the cells and pharmaceutical compositions in ex vivo therapy for the treatment of a disorder (e.g., cancer) in a subject are also provided.

Thus, in one aspect, described herein are cells in which the expression of the B2M gene is modulated (e.g., activated, repressed, or inactivated). In a preferred embodiment, exon 1 or exon 2 of the B2M gene is modulated. Modulation may be by an exogenous molecule (e.g., an engineered transcription factor that contains a DNA binding domain and a transcription activation or repression domain) that binds to the B2M gene and controls B2M expression, and/or through sequence modification of the B2M gene (e.g., using a nuclease that cleaves the B2M gene and modifies the gene sequence by insertion and/or deletion). In certain embodiments, the expression of one or more additional genes is also modulated (e.g., a TCR gene, such as a TCRA gene). In some embodiments, cells are described that contain an engineered nuclease that causes a knockout of the B2M gene such that the class I HLA complex is destabilized. In a preferred embodiment, the destabilization of class I HLA results in a significant reduction of class I HLA complexes on the surface of the cell. In other embodiments, cells are described that include an exogenous molecule, such as an engineered transcription factor (TF), such that the expression of the B2M gene is modulated. In some embodiments, the cell is a T cell. Further described are cells in which the expression of the B2M gene is modulated, where the cell is further engineered to include at least one exogenous transgene or an additional knockout of at least one endogenous gene, or a combination thereof. The exogenous transgene may be integrated into the TCR gene (e.g., when the TCR gene is knocked out), the B2M gene, and/or a non-TCR, non-B2M locus, such as a safe harbor locus. In some cases, the exogenous transgene encodes an ACTR, an engineered TCR, and/or a CAR. The transgene construct may be inserted by a process driven by HDR or NHEJ. In some aspects, cells in which B2M expression is modulated lack key class I HLA on their cell surface and further comprise at least exogenous ACTR and/or exogenous CAR. Some cells comprising a B2M modulator further comprise a knockout of one or more checkpoint inhibitor genes. In some embodiments, the checkpoint inhibitor is PD1. In other embodiments, the checkpoint inhibitor is CTLA4. In further aspects, the B2M modulated cells comprise a PD1 knockout and a CTLA4 knockout. In some embodiments, the cells are further modified with a TCR gene. In some embodiments, the TCR gene that is modulated is the gene encoding TCRβ (TCRB). In some embodiments, this is achieved through targeted truncation of the constant region of this gene (the TCRβ constant region, or TRBC). In certain embodiments, the TCR gene that is modulated is the gene encoding TCRα (TCRA). In further embodiments, the insertion is achieved via targeted truncation of the constant region of the TCR gene, including targeted truncation of the constant region of the TCR alpha gene (referred to herein as the "TRAC" sequence). In some embodiments, the B2M genetically modified cells are further modified with a TCR gene, an HLA-A, -B, -C gene, or a TAP gene, or any combination thereof. In other embodiments, the regulator for HLA class II, CTIIA, is also modified.

In certain embodiments, the cells described herein comprise a modification to the B2M gene (e.g., a modification of exon 1 or exon 2) (e.g., a deletion and/or insertion, binding of an engineered TF to inhibit B2M). In certain embodiments, the modification is a modification of SEQ ID NOs: 6-48 and/or 137-205, including modifications by binding, truncation, insertion and/or deletion of one or more nucleotides within any of these sequences and/or within 1-50 base pairs (including any value therebetween, such as 1-5, 1-10 or 1-20 base pairs) of the genetic (genomic) sequences adjacent to these sequences in the B2M gene. In certain embodiments, the cells comprise a modification in (binding to, truncation, insertion and/or deletion of) one or more of the following sequences within the B2M gene (e.g., exon 1 and/or exon 2, see FIG. 1): GGCCTTA, TCAAATT, TCAAAT, TTACTGA and/or AATTGAA. In certain embodiments, the modification comprises binding of an engineered TF as described herein such that B2M expression is modulated, e.g., repressed or activated. In other embodiments, the modification is a genetic modification (change in nucleotide sequence) at or near the nuclease(s) binding (target) and/or cleavage site(s), including, but not limited to, modifications to the sequence within 1-300 (or any number of base pairs therebetween) base pairs including, downstream and/or one or more base pairs of the cleavage and/or binding site(s); modifications including and/or within 1-100 base pairs (or any number of base pairs therebetween) on either side of the binding and/or cleavage site(s); modifications including and/or within 1-50 base pairs (or any number of base pairs therebetween) on either side (e.g., 1 to 5, 1 to 10, 1 to 20 or more base pairs) of the binding and/or cleavage site(s); and/or modifications to one or more base pairs within the nuclease binding and/or cleavage site. In certain embodiments, the modification is in or near the B2M gene sequence surrounding any of SEQ ID NOs: 6-48 or 137-205 (e.g., 1-300, 1-50, 1-20, 1-10 or 1-5 or more) base pairs, or any number of base pairs therebetween). In certain embodiments, the modification comprises a modification of the B2M gene within one or more of the sequences set forth in SEQ ID NOs: 6-48 or 137-205, or within GGCCTTA, TCAAATT, TCAAAT, TTACTGA and/or AATTGAA (e.g., exon 1 and/or exon 2) of the B2M gene, such as a modification of one or more base pairs to one or more of these sequences. In certain embodiments, the nuclease-mediated gene modification is between paired target sites (when a dimer is used to cleave the target). Nuclease-mediated genetic modifications can include insertions of non-coding sequences of any length and/or transgenes of any length, and/or insertions and/or deletions of any number of base pairs, including deletions of 1 base pair to over 1000 kb (or any value therebetween, including but not limited to 1-100 base pairs, 1-50 base pairs, 1-30 base pairs, 1-20, 1-10, or 1-5 base pairs).

The modified cells of the invention may be lymphoid cells (e.g., T cells), stem/progenitor cells (e.g., induced pluripotent stem cells (iPSCs), embryonic stem cells (e.g., human ES), mesenchymal stem cells (MSCs) or hematopoietic stem cells (HSCs). Stem cells may be totipotent or multipotent (e.g., partially differentiated, such as HSCs, which are multipotent myeloid or lymphoid stem cells). In other embodiments, the invention provides methods for producing cells having a null phenotype for HLA expression. Any of the modified stem cells described herein (modified at the B2M locus) may then be differentiated to generate differentiated (in vivo or in vitro) cells derived from the stem cells described herein. Any of the modified stem cells described herein may include additional modifications in other genes of interest (e.g., TCRA, TCRB, PD1, CTLA4, etc.).

In another aspect, the compositions (modified cells) and methods described herein can be used, for example, in the treatment or prevention or amelioration of a disorder. Typically, the method includes (a) cleaving or downregulating the endogenous B2M gene in an isolated cell (e.g., a T cell or lymphocyte) using a nuclease (e.g., ZFN or TALEN) or a nuclease system such as CRISPR/Cas or Cfp1/CRISPR using engineered crRNA/tracr RNA, or using engineered transcription factors (e.g., ZFN-TF, TALE-TF, Cfp1-TF, or Cas9-TF) such that the B2M gene is inactivated or downmodulated; and (b) introducing the cell into a subject, thereby treating or preventing the disorder.

In some embodiments, the genes encoding TCRα (TCRA) and/or TCRβ (TCRB) are also inactivated or downmodulated. In some embodiments, the inactivation is achieved via targeted cleavage of the constant region of the gene (TCRβ constant region, or TRBC). In a preferred embodiment, the gene encoding TCRα (TCRA) is inactivated or downmodulated. In a further preferred embodiment, the disorder is cancer or an infectious disease. In a further preferred embodiment, the inactivation is achieved via targeted cleavage of the constant region of the gene (TCRα constant region, abbreviated as TRAC).

The transcription factor and/or nuclease(s) can be introduced into the cell as mRNA, in protein form and/or as a DNA sequence encoding the nuclease(s). In certain embodiments, the isolated cells introduced into the subject further comprise additional genomic modifications, such as integrated exogenous sequences (truncated B2M, TCR genes or other genes, e.g., in safe harbor genes or loci) and/or inactivation (e.g., nuclease-mediated) of additional genes, such as one or more HLA and/or TAP genes. The exogenous sequences can be introduced via vectors (e.g., Ad, AAV, LV) or by using techniques such as electroporation. In some embodiments, the proteins are introduced into the cells by cell squeezing (see Kollmannsperger et al. (2016) Nat Comm 7, 10372 doi:10.1038/ncomms10372). In some embodiments, the compositions may include isolated cell fragments and/or differentiated (partially or fully) cells.

In some aspects, the mature cells can be used for cell therapy, e.g., adoptive cell transfer. In other embodiments, cells for use in T cell transplantation contain another genetic modification of interest. In one aspect, the T cells contain an inserted chimeric antigen receptor (CAR) specific for a cancer marker. In a further aspect, the inserted CAR is specific for the CD19 marker characteristic of B cell malignancies. Such cells are useful in therapeutic compositions for treating patients who do not have a matching HLA, and thus can be used as an "off the shelf" therapeutic for any patient in need thereof.

In another aspect, the B2M modulated (modified) T cells contain an inserted antibody-binding T cell receptor (ACTR) donor sequence. In some embodiments, the ACTR donor sequence is inserted into a B2M or TCR gene to disrupt expression of that gene following nuclease-induced cleavage. In other embodiments, the donor sequence is inserted into a "safe harbor" locus, such as the AAVS1, HPRT, albumin, and CCR5 genes. In some embodiments, the ACTR sequence is inserted via targeted integration, in which the ACTR donor sequence contains adjacent homology arms that have homology to sequences adjacent to the engineered nuclease cleavage site. In some embodiments, the ACTR donor sequence further comprises a promoter and/or other transcription control sequences. In other embodiments, the ACTR donor sequence lacks a promoter. In some embodiments, the ACTR donor is inserted into a TCRβ-encoding gene (TCRB). In some embodiments, the insertion is achieved through targeted cleavage of the constant region of this gene (TCRβ constant region or TRBC). In a preferred embodiment, the ACTR donor is inserted into the TCRα coding gene (TCRA). In a further preferred embodiment, the insertion is achieved through targeted cleavage of the constant region of this gene (TCRα constant region, abbreviated as TRAC). In some embodiments, the donor is inserted into an exonic sequence in the TCRA, while in others the donor is inserted into an intronic sequence in the TCRA. In some embodiments, the TCR modulated cell further comprises a CAR. In yet a further embodiment, the B2M modulated cell is further modulated with an HLA gene or a checkpoint inhibitor gene.

Also provided are pharmaceutical compositions comprising modified cells (e.g., T cells or stem cells with an inactivated B2M gene) as described herein, or one or more B2M binding molecules (e.g., engineered transcription factors and/or nucleases) as described herein. In certain embodiments, the pharmaceutical compositions further comprise one or more pharma- ceutical acceptable excipients. The modified cells, B2M binding molecules (or polynucleotides encoding these molecules) and/or pharmaceutical compositions comprising these cells or molecules are introduced into the subject via methods known in the art, for example, via intravenous infusion, injection into certain blood vessels such as the hepatic artery, and via direct tissue injection (e.g., muscle). In some embodiments, the subject is an adult with a disease or condition that can be treated or ameliorated with the composition. In other embodiments, the subject is a pediatric subject to whom the composition is administered to prevent, treat, or ameliorate a disease or condition (e.g., cancer, graft-versus-host disease, etc.).

In some aspects, the composition (ACTR-containing B2M modulated cells) further comprises an exogenous antibody. See U.S. Patent Application No. 15/357,772. In some aspects, the antibody is useful for preventing or treating a condition comprising T cells that contain ACTR. In some embodiments, the antibody recognizes an antigen associated with tumor cells or associated with cancer-related processes, such as EpCAM, CEA, gpA33, mucin, TAG-72, CAIX, PSMA, folate binding antibodies, CD19, EGFR, ERBB2, ERBB3, MET, IGF1R, EPHA3, TRAILR1, TRAILR2, RANKL, FAP, VEGF, VEGFR, αVβ3 and α5β1 integrins, CD20, CD30, CD33, CD52, CTLA4, and enascin (Scott et al. (2012) Nat Rev Cancer 12:278). In other embodiments, the antibody recognizes an antigen associated with an infectious disease, such as HIV, HCV, and the like.

In another aspect, provided herein are B2M DNA binding domains (e.g., ZFPs, TALEs, and sgRNAs) that bind to target sites in the B2M gene. In certain embodiments, the DNA binding domain comprises a ZFP having recognition helix regions in the order shown in a row of Table 1; a TAL effector domain DNA binding protein having an RVD shown in a row of Table 2; and/or a sgRNA shown in a row of Table 2A. These DNA binding proteins can associate with a transcriptional regulatory domain to form an engineered transcription factor that modulates B2M expression. Alternatively, these DNA binding proteins can associate with one or more nuclease domains to form engineered zinc finger nucleases (ZFNs), TALENs, and/or CRISPR/Cas systems that bind and cleave the B2M gene. In certain embodiments, a single guide RNA (sgRNA) of a ZFN, TALEN, or CRISPR/Cas system binds to a target site in the human B2M gene. A DNA binding domain of a transcription factor or nuclease (e.g., ZFP, TALE, sgRNA) can bind to a target site in the B2M gene that includes 9, 10, 11 12 or more (e.g., 13, 14, 15, 16, 17, 18, 19, 20 or more) nucleotides of any of SEQ ID NOs: 6-48 or 137-205. A zinc finger protein may include one, two, three, four, five, six or more zinc fingers, each of which has a recognition helix that specifically contacts a target subsite in the target gene. In certain embodiments, the zinc finger protein comprises four or five or six fingers (designated F1, F2, F3, F4, F5 and F6, ordered N-terminus to C-terminus from F1 to F4 or F5 or F6), e.g., as shown in Table 1. In other embodiments, a single guide RNA or TAL effector DNA binding domain can bind to a target site set forth in any of SEQ ID NOs: 6-48 or 137-205 (or 12 or more base pairs within any of SEQ ID NOs: 6-48). Exemplary sgRNA target sites are set forth in SEQ ID NOs: 16-48, and exemplary TALEN binding sites are set forth in Table 2B (SEQ ID NOs: 137-205). Additional TALENs can be designed to target the sites described herein using canonical or non-canonical RVDs as described in U.S. Pat. Nos. 8,586,526 and 9,458,205. The nucleases described herein (including ZFP, TALE or sgRNA DNA binding domains) are capable of making genetic modifications in the B2M gene, including any of SEQ ID NOs: 6-48 or 137-205, including modifications (insertions and/or deletions) within any of these sequences (SEQ ID NOs: 6-48 or 137-205) and/or modifications to the B2M gene sequence adjacent to the target site sequence set forth in SEQ ID NOs: 6-48 or 137-205, such as modifications within exon 1/or exon 2 of the B2M gene within one or more of the following sequences: GGCCTTA, TCAAATT, TCAAAT, TTACTGA and/or AATTGAA.

Also provided is a fusion molecule comprising a DNA binding domain that binds to exon 1 or exon 2 of the B2M gene and a transcriptional regulatory domain or a nuclease domain, wherein the DNA binding domain comprises a zinc finger protein (ZFP) shown in a row of Table 1, a TALE effector protein shown in a row of Table 2B, or a single guide RNA (sgRNA) shown in a row of Table 2A.

Any of the proteins described herein can further comprise a cleavage domain and/or cleavage half-domain (e.g., a wild-type or engineered FokI cleavage half-domain). Thus, in any of the nucleases described herein (e.g., ZFN, TALEN, CRISPR/Cas systems), the nuclease domain can comprise a wild-type nuclease domain or nuclease half-domain (e.g., a FokI cleavage half-domain). In other embodiments, the nuclease (e.g., ZFN, TALEN, CRISPR/Cas nuclease) comprises an engineered nuclease domain or half-domain, e.g., an engineered FokI cleavage half-domain that forms an obligate heterodimer. See, e.g., U.S. Patent Application Publication No. 20080131962.

In another aspect, the disclosure provides a polynucleotide (e.g., sgRNA or other DNA binding domain) encoding any of the proteins, fusion molecules and/or components thereof described herein. The polynucleotide may be part of a viral vector, a non-viral vector (e.g., a plasmid) or in mRNA form. Any of the polynucleotides described herein may include sequences (donor, homology arm or patch sequences) for targeted insertion into the B2M, TCRα and/or TCRβ genes. In yet another aspect, a gene delivery vector is provided that includes any of the polynucleotides described herein. In certain embodiments, the vector is an adenoviral vector (e.g., an Ad5/F35 vector) or a lentiviral vector (LV) or adeno-associated vector (AAV), including integration-competent or integration-deficient lentiviral vectors. Thus, also provided herein are viral vectors comprising sequences encoding a nuclease (e.g., ZFN or TALEN) and/or a nuclease system (CRISPR/Cas or Ttago) and/or a donor sequence for targeted integration into a target gene. In some embodiments, the donor sequence and the sequence encoding the nuclease are on different vectors. In other embodiments, the nuclease is provided as a polypeptide. In a preferred embodiment, the polynucleotide is an mRNA. In some aspects, the mRNA may be chemically modified (see, e.g., Kormann et al., (2011) Nature Biotechnology 29(2):154-157). In other aspects, the mRNA may include an ARCA cap (see, U.S. Pat. Nos. 7,074,596 and 8,153,773). In some aspects, the mRNA may include a cap introduced by enzymatic modification. The enzymatically introduced cap may include Cap0, Cap1, or Cap2 (see, e.g., Smietanski et al. (2014) Nature Communications 5:3004). In further aspects, the mRNA may be capped by chemical modification. In further embodiments, the mRNA may include a mixture of unmodified and modified nucleotides (see, U.S. Patent Application Publication No. 2012-0195936). In yet further embodiments, the mRNA may include a WPRE element (see, U.S. Patent Application No. 15/141,333). In some embodiments, the mRNA is double-stranded (see, e.g., Kariko et al. (2011) Nucl Acid Res 39:e142).

In yet another aspect, the disclosure provides isolated cells comprising any of the proteins, polynucleotides and/or vectors described herein. In certain embodiments, the cells are selected from the group consisting of stem/progenitor cells, or T cells (e.g., CD4 ⁺ T cells). In yet a further aspect, the disclosure provides cells or cell lines derived from cells or cell lines comprising any of the proteins, polynucleotides and/or vectors described herein, i.e., cells or cell lines derived (e.g., in culture) from cells in which B2M has been inactivated by one or more ZFNs and/or a donor polynucleotide (e.g., ACTR, engineered TCR and/or CAR) has been stably integrated into the genome of the cells. Thus, progeny of the cells described herein may not themselves comprise the proteins, polynucleotides and/or vectors described herein, but in these cells at least one B2M gene has been inactivated and/or a donor polynucleotide has been integrated into the genome and/or is expressed.

In another aspect, described herein are methods of inactivating the B2M gene in a cell by introducing one or more proteins, polynucleotides and/or vectors described herein into the cell. In any of the methods described herein, the nuclease can induce targeted mutagenesis, deletion of cellular DNA sequences, and/or promote targeted recombination at a given chromosomal locus, whereby in certain embodiments the nuclease deletes one or more nucleotides from or inserts one or more nucleotides into the target gene. In some embodiments, the B2M gene is inactivated by nuclease cleavage followed by non-homologous end joining. In other embodiments, the genomic sequence in the target gene is replaced using, for example, a nuclease (or a vector encoding said nuclease) as described herein and a "donor" sequence that is inserted into the gene following targeted cleavage by the nuclease. The donor sequence can be present in the nuclease vector, can be present in a separate vector (e.g., AAV, Ad or LV vector), or can be introduced into the cell using a different nucleic acid delivery mechanism.

Additionally, any of the methods described herein can be performed in vitro, in vivo, and/or ex vivo. In certain embodiments, the methods are performed ex vivo, for example to modify T cells to make them useful as therapeutic agents in an allogenic setting to treat a subject (e.g., a subject with cancer). Non-limiting examples of cancers that can be treated and/or prevented include lung cancer, pancreatic cancer, liver cancer, bone cancer, breast cancer, colorectal cancer, leukemia, ovarian cancer, lymphoma, brain cancer, and the like.
In certain embodiments, for example, the following are provided:
(Item 1)
An isolated cell, in which expression of the beta 2 microglobulin (B2M) gene is modulated by modification of exon 1 or exon 2 of the B2M gene.
(Item 2)
The cell of item 1, wherein the modification is produced by an exogenous fusion molecule comprising a DNA binding domain and a functional domain that binds to a target site set forth in any of SEQ ID NOs: 6-48 or 137-205.
(Item 3)
2. The cell of item 1, comprising an insertion and/or deletion within one or more of SEQ ID NOs: 6-48, or 137-205; within 1 to 10 base pairs on either side (flanking genomic sequences) of SEQ ID NOs: 6-48, or 137-205; or within GGCCTTA, TCAAAT, TCAAATT, TTACTGA, and/or AATTGAA.
(Item 4)
4. The cell of any of items 1 to 3, further comprising an inactive T cell receptor gene, a PD1 and/or a CTLA4 gene.
(Item 5)
5. The cell of any of items 1 to 4, further comprising a transgene encoding a chimeric antigen receptor (CAR), a transgene encoding an antibody-binding T cell receptor (ACTR), and/or a transgene encoding an engineered TCR.
(Item 6)
6. The cell according to any one of items 1 to 5, which is a lymphoid cell, a stem cell or a progenitor cell.
(Item 7)
7. The cell of item 6, which is a T cell, an induced pluripotent stem cell (iPSC), an embryonic stem cell, a mesenchymal stem cell (MSC) or a hematopoietic stem cell (HSC).
(Item 8)
A pharmaceutical composition comprising the cell according to any one of items 1 to 7.
(Item 9)
A fusion molecule comprising a DNA binding domain that binds to exon 1 or exon 2 of the B2M gene and a transcriptional regulatory domain or a nuclease domain, wherein the DNA binding domain comprises a zinc finger protein (ZFP) shown in a row of Table 1, a TALE effector protein shown in a row of Table 2B, or a single guide RNA (sgRNA) shown in a row of Table 2A.
(Item 10)
A polynucleotide encoding the fusion molecule of item 9.
(Item 11)
11. The polynucleotide of item 10, which is a viral vector, a plasmid or an mRNA.
(Item 12)
A method for treating or preventing cancer, comprising introducing a cell according to any one of items 1 to 7 or a pharmaceutical composition according to item 8 into a subject having cancer.
(Item 13)
13. Use of the cell according to any one of items 1 to 7, the pharmaceutical composition according to item 8, or the fusion molecule according to item 9, or the polynucleotide according to item 10, for treating a subject with cancer.

1 is a diagram of the sequences of exon 1 and exon 2 of the B2M gene targeted by the nuclease (SEQ ID NOs: 1 and 2). Boxes indicate five different cleavage regions (A (GGCCTTA), C (TCAAATT), D (TCAAAT), E (TTACTGA) and G (AATTGAA)) flanked by ZFN binding (target) sites.

Figures 2A and 2B show nuclease activity. Figure 2A is a bar graph showing the percentage of gene modification at each site in T cells treated with ZFNs specific for B2M sites A, C, D, E and G shown in Figure 1 at a dose of 2 or 6 μg. Figure 2B shows TALEN activity against the B2M gene in K562 cells. Figures 2A and 2B show nuclease activity. Figure 2A is a bar graph showing the percentage of gene modification at each site in T cells treated with ZFNs specific for B2M sites A, C, D, E and G shown in Figure 1 at a dose of 2 or 6 μg. Figure 2B shows TALEN activity against the B2M gene in K562 cells.

FIG. 3 shows the percentage of HLA-negative T cells following treatment with B2M-specific ZFN pairs as analyzed by FACS analysis.

Figures 4A to 4E show FACS results from treatment of cells with both B2M and TCRA specific ZFNs. Figure 4A shows the results without ZFN treatment. Figure 4B shows the results after TCRA specific ZFN alone (96% knockout of CD3 signal) and Figure 4C shows the results after B2M specific ZFN alone (92% knockout of HLA signal). Figure 4D is an illustration showing the location of cells with a double knockout (resulting in a reduction of both HLA and CD3 marking). Figure 4E shows the results after treatment of cells with both TCRA specific ZFN and CD3 specific ZFN, demonstrating a double knockout in 82% of cells. Figures 4A to 4E show FACS results from treatment of cells with both B2M and TCRA specific ZFNs. Figure 4A shows the results without ZFN treatment. Figure 4B shows the results after TCRA specific ZFN alone (96% knockout of CD3 signal) and Figure 4C shows the results after B2M specific ZFN alone (92% knockout of HLA signal). Figure 4D is an illustration showing the location of cells with a double knockout (resulting in a reduction of both HLA and CD3 marking). Figure 4E shows the results after treatment of cells with both TCRA specific ZFN and CD3 specific ZFN, demonstrating a double knockout in 82% of cells.

FIG. 5 shows results from a TRAC (TCRA) and B2M double knockout and targeted integration of the donor into either the TRAC (TCRA) or B2M locus.

DETAILED DESCRIPTION Disclosed herein are compositions and methods for generating cells in which expression of the B2M gene has been modulated such that they no longer contain HLA class I on the cell surface. Cells modified in this manner can be used as therapeutics, e.g., grafts, since the lack of B2M expression prevents or reduces HLA-based immune responses. Additionally, other genes of interest can be inserted into cells in which the B2M gene has been engineered, and/or other genes of interest can be knocked out.

General The practice of the methods disclosed herein, and the preparation and use of the compositions, employ, unless otherwise indicated, conventional techniques of molecular biology, biochemistry, chromatin structure and analysis, computational chemistry, cell culture, recombinant DNA and related fields that are within the skill of the art. These techniques are explained at length in the literature. See, e.g., Sambrook et al., MOLECULAR CLONING: A LABORATORY MANUAL, 2nd ed., Cold Spring Harbor Laboratory Press, 1989 and 3rd ed., 2001; Ausubel et al., CURRENT PROTOCOLS IN MOLECULAR BIOLOGY, John Wiley & Sons, New York, 1987 and periodic updates; METHODS IN ENZYMOLOGY series, Academic Press, San Diego; Wolffe, CHROMATIN STRUCTURE AND FUNCTION, 3rd ed., Academic Press, San Diego, 1998; METHODS IN ENZYMOLOGY, Vol. 304, "Chromatin" (eds. P. M. Wassarman and A. P. Wolffe), Academic Press, San Diego, 1999; and METHODS IN MOLECULAR BIOLOGY, Vol. 119, "Chromatin Protocols" (ed. P. B. Becker), Humana Press, Totowa, 1999.

Definitions The terms "nucleic acid", "polynucleotide" and "oligonucleotide" are used interchangeably and refer to deoxyribonucleotide or ribonucleotide polymers in linear or circular conformation and in single- or double-stranded form. For purposes of this disclosure, these terms should not be construed as limiting with respect to the length of the polymer. These terms can encompass known analogs of natural nucleotides as well as nucleotides in which the base, sugar and/or phosphate moieties (e.g., phosphorothioate backbones) are modified. In general, analogs of a particular nucleotide have the same base-pairing specificity; i.e., an analog of A will base-pair with T.

The terms "polypeptide," "peptide," and "protein" are used interchangeably to refer to a polymer of amino acid residues. The terms also apply to amino acid polymers in which one or more amino acids are chemical analogues or modified derivatives of a corresponding naturally occurring amino acid.

"Binding" refers to a sequence-specific, non-covalent interaction between macromolecules (e.g., between a protein and a nucleic acid). Not all components of a binding interaction need be sequence-specific (e.g., contacts with phosphate residues in a DNA backbone), as long as the binding interaction as a whole is sequence-specific. Such interactions are generally characterized by a dissociation constant (K _d ) of 10 ⁻⁶ M ⁻¹ or lower. "Affinity" refers to the strength of binding. Increased binding affinity correlates with a lower K _d . "Non-specific binding" refers to a target sequence-independent, non-covalent interaction that occurs between any molecule of interest (e.g., an engineered nuclease) and a macromolecule (e.g., DNA).

A "DNA binding molecule" is a molecule that can bind to DNA. Such a DNA binding molecule may be a polypeptide, a domain of a protein, a domain within a larger protein, or a polynucleotide. In some embodiments, the polynucleotide is DNA, while in other embodiments, the polynucleotide is RNA. In some embodiments, the DNA binding molecule is a protein domain of a nuclease (e.g., a FokI domain), while in other embodiments, the DNA binding molecule is a guide RNA component of an RNA-guided nuclease (e.g., Cas9 or Cfp1).

A "binding protein" is a protein that can bind non-covalently to another molecule. A binding protein can bind, for example, to a DNA molecule (a DNA-binding protein), an RNA molecule (an RNA-binding protein), and/or a protein molecule (a protein-binding protein). In the case of a protein-binding protein, it can bind to itself (forming homodimers, homotrimers, etc.) and/or it can bind to one or more molecules of one or more different proteins. A binding protein can have more than one type of binding activity. For example, a zinc finger protein has DNA-binding activity, RNA-binding activity, and protein-binding activity.

A "zinc finger DNA-binding protein" (or binding domain) is a protein, or a domain within a larger protein, that binds to DNA in a sequence-specific manner through one or more zinc fingers, which are regions of amino acid sequence within the binding domain whose structure is stabilized through the coordination of a zinc ion. The term zinc finger DNA-binding protein is often abbreviated as zinc finger protein or ZFP.
A "TALE DNA binding domain" or "TALE" is a polypeptide that contains one or more TALE repeat domains/units. Each repeat domain contains a repeat variable diresidue (RVD) and is responsible for binding of the TALE to its cognate target DNA sequence. A single "repeat unit" (also called a "repeat") is typically 33-35 amino acids in length and exhibits at least some sequence homology with other TALE repeat sequences among naturally occurring TALE proteins. TALE proteins can be designed to bind to target sites using canonical or non-canonical RVDs within the repeat unit. See, for example, U.S. Patent Nos. 8,586,526 and 9,458,205, which are incorporated by reference herein in their entireties.

Zinc finger and TALE DNA binding domains can be "engineered" to bind to a given nucleotide sequence, for example, through engineering of the recognition helix region of a naturally occurring zinc finger protein (changing one or more amino acids) or by engineering of amino acids involved in DNA binding (repeated variable dinucleotide or RVD regions). Thus, engineered zinc finger or TALE proteins are non-naturally occurring proteins. A non-limiting example of a method for engineering zinc finger proteins and TALEs is design and selection. Designed proteins are non-naturally occurring proteins whose design/composition arises primarily from rational criteria. Rational criteria for design include application of substitution rules and computerized algorithms to process information in databases that store information on existing ZFP or TALE designs (canonical and non-canonical RVDs) and binding data. See, for example, U.S. Patent Nos. 9,458,205, 8,586,526, 6,140,081, 6,453,242 and 6,534,261, and also WO 98/53058, WO 98/53059, WO 98/53060, WO 02/016536 and WO 03/016496.

The "selected" zinc finger proteins, TALE proteins, or CRISPR/Cas systems are not found in nature and their production arises primarily from experimental processes such as phage display, interaction traps, or hybrid selection. See, e.g., U.S. 5,789,538; U.S. 5,925,523; U.S. 6,007,988; U.S. 6,013,453; U.S. 6,200,759; WO 95/19431; WO 96/06166; WO 98/53057; WO 98/54311; WO 00/27878; WO 01/60970; WO 01/88197, and WO 02/099084.

"TtAgo" is a prokaryotic Argonaute protein believed to be involved in gene silencing. TtAgo is derived from the bacterium Thermus thermophilus. See, e.g., Swarts et al., ibid.; G. Sheng et al., (2013) Proc. Natl. Acad. Sci. U.S.A. 111, 652). A "TtAgo system" is all of the required components, including, e.g., guide DNA for cleavage by the TtAgo enzyme.

"Recombination" refers to the process of exchange of genetic information between two polynucleotides. For purposes of this disclosure, "homologous recombination (HR)" refers to a specialized form of such exchange that occurs, for example, during repair of double-strand breaks in cells by the homology-directed repair mechanism. This process requires nucleotide sequence homology, uses a "donor" molecule to template repair of a "target" molecule (i.e., the one that underwent the double-strand break), and is variously known as "non-crossover gene conversion" or "short-tract gene conversion" because it results in the transfer of genetic information from the donor to the target. Without wishing to be bound by any particular theory, such transfer can involve mismatch correction of the heteroduplex DNA formed between the cleaved target and the donor, and/or "synthesis-dependent strand annealing," and/or related processes used by the donor to resynthesize the genetic information that becomes part of the target. Such specialized HR often results in alterations of the sequence of the target molecule, such that part or all of the sequence of the donor polynucleotide is incorporated into the target polynucleotide.

In the disclosed methods, one or more targeted nucleases described herein create a double-stranded break (DSB) in a target sequence (e.g., cellular chromatin) at a predetermined site (e.g., a gene or locus of interest), and a "donor" polynucleotide having homology to a nucleotide sequence in the region of the break can be introduced into the cell. The presence of a DSB has been shown to promote integration of the donor sequence. Optionally, the construct has homology to a nucleotide sequence in the region of the break. The donor sequence may be physically integrated, or alternatively, the donor polynucleotide is used as a template for repair of the break by homologous recombination, resulting in the introduction of all or a portion of the nucleotide sequence as in the donor into the cellular chromatin. Thus, a first sequence in the cellular chromatin can be modified, and in certain embodiments, converted to a sequence present in the donor polynucleotide. Thus, it can be understood that use of the terms "replace" or "replacement" refers to the replacement of one nucleotide sequence with another (i.e., replacement of the sequence in the informational sense) and does not necessarily require the physical or chemical replacement of one polynucleotide with another.

In any of the methods described herein, additional pairs of zinc finger proteins can be used for additional double-stranded cleavage at additional target sites within the cell.

In certain embodiments of the method for targeted recombination and/or replacement and/or alteration of sequences in a region of interest in cellular chromatin, the chromosomal sequence is altered by homologous recombination with an exogenous "donor" nucleotide sequence. Such homologous recombination is stimulated by the presence of a double-stranded break in cellular chromatin, provided that a homologous sequence is present in the region of the break.

In any of the methods described herein, the first nucleotide sequence (the "donor sequence") can contain a sequence that is homologous but not identical to the genomic sequence of the region of interest, thereby stimulating homologous recombination to insert the non-identical sequence into the region of interest. Thus, in certain embodiments, the portion of the donor sequence that is homologous to the sequence of the region of interest exhibits between about 80 and 99% (or any integer value therebetween) sequence identity with the replaced genomic sequence. In other embodiments, the homology between the donor sequence and the genomic sequence is greater than 99%, for example, when only one nucleotide differs between the donor sequence and the genomic sequence over 100 contiguous base pairs. In certain cases, the non-homologous portion of the donor sequence can contain a sequence that is not present in the region of interest, such that new sequence is introduced into the region of interest. In these cases, the non-homologous sequence is generally flanked by 50 to 1,000 base pairs (or any integer value therebetween) or any number of base pairs greater than 1,000 that are homologous or identical to the sequence of the region of interest. In other embodiments, the donor sequence is non-homologous to the first sequence and is inserted into the genome by non-homologous recombination mechanisms.

Any of the methods described herein can be used for partial or complete inactivation of one or more target sequences in a cell by targeted integration of a donor sequence that disrupts expression of the gene(s) of interest. Cell lines with partially or completely inactivated genes are also provided.

Additionally, the targeted integration methods described herein can also be used to integrate one or more exogenous sequences. The exogenous nucleic acid sequence can include, for example, one or more genes or cDNA molecules, or any type of coding or non-coding sequence, as well as one or more regulatory elements (e.g., promoters). Additionally, the exogenous nucleic acid sequence can produce one or more RNA molecules (e.g., small hairpin RNA (shRNA), inhibitory RNA (RNAi), microRNA (miRNA), etc.).

"Cleavage" refers to the cleavage of the covalent backbone of a DNA molecule. Cleavage can be initiated by a variety of methods, including, but not limited to, enzymatic or chemical hydrolysis of phosphodiester bonds. Both single-stranded and double-stranded breaks are possible, with double-stranded breaks occurring as a result of two different single-stranded break events. DNA cleavage can result in the production of either blunt ends or staggered ends. In certain embodiments, fusion polypeptides are used for targeted double-stranded DNA cleavage.

A "cleavage half-domain" is a polypeptide sequence that, together with a second polypeptide (either the same or different), forms a complex having cleavage activity (preferably double-strand cleavage activity). The terms "first and second cleavage half-domains," "+ and - cleavage half-domains," and "right and left cleavage half-domains" are used interchangeably to refer to pairs of cleavage half-domains that dimerize.

An "engineered cleavage half-domain" is a cleavage half-domain that has been modified to form an obligate heterodimer with another cleavage half-domain (e.g., another engineered cleavage half-domain). See also U.S. Pat. Nos. 7,888,121; 7,914,796; 8,034,598; 8,623,618 and U.S. Patent Application Publication No. 2011/0201055, which are incorporated by reference in their entireties.

The term "sequence" refers to a nucleotide sequence of any length, which may be DNA or RNA; may be linear, circular or branched, and may be either single-stranded or double-stranded. The term "donor sequence" refers to a nucleotide sequence that is inserted into a genome. The donor sequence may be of any length, for example, between 2 and 10,000 nucleotides (or any integer value therebetween or therebetween), preferably between about 100 and 1,000 nucleotides (or any integer value therebetween), more preferably between about 200 and 500 nucleotides in length.

"Chromatin" is the nucleoprotein structure that comprises the cellular genome. Cellular chromatin contains nucleic acids, primarily DNA, and proteins, including histones and non-histone chromosomal proteins. Most eukaryotic cellular chromatin exists in the form of nucleosomes, in which the nucleosome core contains approximately 150 base pairs of DNA associated with an octamer containing two each of histones H2A, H2B, H3, and H4; linker DNA (of variable length depending on the organism) extends between the nucleosome cores. A molecule of histone H1 is commonly associated with the linker DNA. For purposes of this disclosure, the term "chromatin" is meant to encompass all types of cellular nucleoproteins, both prokaryotic and eukaryotic. Cellular chromatin includes both chromosomal and episomal chromatin.

A "chromosome" is a chromatin complex that comprises all or part of a cell's genome. The genome of a cell is often characterized by its karyotype, which is the collection of all chromosomes that comprise the genome of the cell. The genome of a cell can include one or more chromosomes.

An "episome" is a replicating nucleic acid, nucleoprotein complex, or other structure that contains nucleic acid that is not part of the chromosomal karyotype of a cell. Examples of episomes include plasmids and certain viral genomes.

A "target site" or "target sequence" is a nucleic acid sequence that defines a portion of a nucleic acid to which a binding molecule binds, provided sufficient conditions for binding exist. For example, the sequence 5'GAATTC3' is a target site for the EcoRI restriction endonuclease.

An "exogenous" molecule is one that is not normally present in a cell, but that can be introduced into a cell by one or more genetic, biochemical, or other methods. "Normal presence in a cell" is determined with respect to the particular developmental stage and environmental conditions of the cell. Thus, for example, a molecule that is present only during embryonic development of muscle is an exogenous molecule with respect to an adult muscle cell. Similarly, a molecule that is induced by heat shock is an exogenous molecule with respect to a non-heat shocked cell. Exogenous molecules can include, for example, a functional version of a dysfunctional endogenous molecule or a dysfunctional version of a normally functioning endogenous molecule.

Exogenous molecules may be, inter alia, small molecules, such as those produced by combinatorial chemical processes, or macromolecules, such as proteins, nucleic acids, carbohydrates, lipids, glycoproteins, lipoproteins, polysaccharides, any modified derivatives of the above molecules, or any complexes containing one or more of the above molecules. Nucleic acids include DNA and RNA, may be single-stranded or double-stranded, may be linear, branched or circular, and may be of any length. See, for example, U.S. Patent Nos. 8,703,489 and 9,255,259. Nucleic acids include nucleic acids capable of forming duplexes as well as triplex-forming nucleic acids. See, for example, U.S. Patent Nos. 5,176,996 and 5,422,251. Proteins include, but are not limited to, DNA binding proteins, transcription factors, chromatin remodeling factors, methylated DNA binding proteins, polymerases, methylases, demethylases, acetylases, deacetylases, kinases, phosphatases, integrases, recombinases, ligases, topoisomerases, gyrases, and helicases.

An exogenous molecule may be the same type of molecule as an endogenous molecule, e.g., an exogenous protein or nucleic acid. For example, an exogenous nucleic acid may include an infectious viral genome, a plasmid or episome introduced into a cell, or a chromosome not normally present in the cell. Methods for the introduction of exogenous molecules into cells are known to those of skill in the art and include, but are not limited to, lipid-mediated transfer (i.e., liposomes containing neutral and cationic lipids), electroporation, direct injection, cell fusion, microprojectile bombardment, calcium phosphate co-precipitation, DEAE-dextran mediated transfer, and viral vector mediated transfer. An exogenous molecule may be the same type of molecule as an endogenous molecule, but may be derived from a different species than the cell is derived from. For example, a human nucleic acid sequence may be introduced into a cell line originally derived from a mouse or hamster.

In contrast, an "endogenous" molecule is one that is normally present in a particular cell at a particular developmental stage under particular environmental conditions. For example, endogenous nucleic acids can include chromosomes, mitochondrial genomes, chloroplasts or other organelles, or naturally occurring episomal nucleic acids. Additional endogenous molecules can include proteins, e.g., transcription factors and enzymes.

A "fusion" molecule is a molecule in which two or more subunit molecules are linked, preferably covalently. The subunit molecules may be of the same chemical type or may be of different chemical types. Examples of the first type of fusion molecule include, but are not limited to, fusion proteins (e.g., fusions between a ZFP or TALE DNA binding domain and one or more activation domains) and fusion nucleic acids (e.g., nucleic acids encoding the fusion proteins described above). Examples of the second type of fusion molecule include, but are not limited to, fusions between triplex-forming nucleic acids and polypeptides, and fusions between minor groove binders and nucleic acids. The term also includes systems in which a polynucleotide component associates with a polypeptide component to form a functional molecule (e.g., the CRISPR/Cas system in which a single guide RNA associates with a functional domain to modulate gene expression).

Expression of a fusion protein in a cell can result from delivery of the fusion protein to the cell or by delivery of a polynucleotide encoding the fusion protein to the cell, where the polynucleotide is transcribed and the transcript is translated to produce the fusion protein. Expression of the protein in a cell can also involve trans-splicing, polypeptide cleavage and polypeptide ligation. Methods of delivery of polynucleotides and polypeptides to cells are presented elsewhere in this disclosure.

For purposes of this disclosure, a "gene" includes a DNA region that encodes a gene product (see below), as well as all DNA regions that regulate the production of the gene product, whether or not such regulatory sequences are adjacent to the coding and/or transcribed sequence. Thus, a gene includes, but is not necessarily limited to, promoter sequences, terminators, translational regulatory sequences, such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators, boundary elements, origins of replication, matrix attachment sites, and locus control regions.

A "safe harbor" locus is a locus in the genome where a gene can be inserted without any deleterious effects on the host cell. Most beneficial are safe harbor loci where the expression of the inserted gene sequence is not perturbed by any read-through expression from nearby genes. Non-limiting examples of safe harbor loci targeted by nuclease(s) include CCR5, CCR5, HPRT, AAVS1, Rosa and albumin. See, for example, U.S. Patent Nos. 8,771,985; 8,110,379; 7,951,925; U.S. Patent Application Publication Nos. 20100218264; 20110265198; 20130137104; 20130122591; 20130177983; 20130177960; 20150056705 and 20150159172).

"Gene expression" refers to the conversion of the information contained in a gene into a gene product. A gene product may be the direct transcription product of a gene (e.g., mRNA, tRNA, rRNA, antisense RNA, ribozyme, structural RNA or any other type of RNA) or a protein produced by translation of an mRNA. Gene products also include RNA that is modified by processes such as capping, polyadenylation, methylation and editing, as well as proteins that are modified, for example, by methylation, acetylation, phosphorylation, ubiquitination, ADP-ribosylation, myristoylation and glycosylation.

"Modulation" or "alteration" of gene expression refers to a change in the activity of a gene. Modulation of expression can include, but is not limited to, gene activation, including by modification of the gene via binding of an exogenous molecule (e.g., an engineered transcription factor), and gene repression. Modulation can also be achieved by alteration of the gene sequence via genome editing (e.g., truncation, alteration, inactivation, random mutation). Gene inactivation refers to any reduction in gene expression as compared to an unmodified cell as described herein. Thus, gene inactivation may be partial or complete.

A "region of interest" is any region of cellular chromatin, such as, for example, a gene or a non-coding sequence within or adjacent to a gene, to which it is desirable to bind an exogenous molecule. Binding may be for targeted DNA cleavage and/or targeted recombination. A region of interest can be present, for example, in a chromosome, an episome, an organelle genome (e.g., mitochondria, chloroplasts), or, for example, in an infectious viral genome. A region of interest may be within the coding region of a gene, within a transcribed non-coding region, such as, for example, a leader sequence, trailer sequence, or intron, or within a non-transcribed region either upstream or downstream of a coding region. A region of interest may be as small as a single nucleotide pair or may be up to 2,000 nucleotide pairs in length, or any integer value of nucleotide pairs.

"Eukaryotic" cells include, but are not limited to, fungal cells (such as yeast), plant cells, animal cells, mammalian cells, and human cells (e.g., T cells).

The terms "operably linked" and "operably linked" (or "operably linked") are used interchangeably in reference to the juxtaposition of two or more components (such as sequence elements) where the components are positioned to allow both components to function normally such that at least one of the components can mediate a function expressed in at least one of the other components. By way of illustration, a transcriptional regulatory sequence such as a promoter is operably linked to a coding sequence if the transcriptional regulatory sequence regulates the level of transcription of the coding sequence in response to the presence or absence of one or more transcriptional regulatory factors. A transcriptional regulatory sequence is generally operably linked in cis to a coding sequence, but need not be directly adjacent to it. For example, an enhancer is a transcriptional regulatory sequence that is operably linked to a coding sequence, even if they are not contiguous.

With respect to fusion polypeptides, the term "operably linked" can refer to the fact that each of the components performs the same function in conjunction with the other component that it would if not so linked. For example, with respect to a fusion polypeptide in which a DNA-binding domain (e.g., ZFP, TALE) is fused to an activation domain, the DNA-binding domain and the activation domain are operably linked if, in the fusion polypeptide, the DNA-binding domain portion can bind to its target site and/or its binding site, and the activation domain can upregulate gene expression. In the case of a fusion polypeptide in which a DNA-binding domain is fused to a cleavage domain, the DNA-binding domain and the cleavage domain are operably linked if, in the fusion polypeptide, the DNA-binding domain portion can bind to its target site and/or its binding site, and the cleavage domain can cleave DNA near the target site. Similarly, for a fusion polypeptide in which a DNA-binding domain is fused to an activation or repression domain, the DNA-binding domain and the activation or repression domain are in operative linkage if the DNA-binding domain portion is capable of binding to its target site and/or its binding site in the fusion polypeptide, while the activation domain is capable of upregulating gene expression or the repression domain is capable of downregulating gene expression.

A "functional fragment" of a protein, polypeptide, or nucleic acid is a protein, polypeptide, or nucleic acid whose sequence is not identical to the full-length protein, polypeptide, or nucleic acid, but which retains the same function as the full-length protein, polypeptide, or nucleic acid. A functional fragment can possess more, fewer, or the same number of residues as the corresponding native molecule, and/or can contain one or more amino acid or nucleotide substitutions. Methods for determining the function of a nucleic acid (e.g., coding function, ability to hybridize to another nucleic acid) are well known in the art. Similarly, methods for determining protein function are well known. For example, DNA binding function of a polypeptide can be determined, for example, by filter binding, electrophoretic mobility-shift, or immunoprecipitation assays. DNA cleavage can be tested by gel electrophoresis. See Ausubel et al., supra. The ability of a protein to interact with another protein can be determined, for example, by co-immunoprecipitation, two-hybrid assays, or both genetic and biochemical complementation. See, e.g., Fields et al. (1989) Nature 340:245-246; U.S. Patent No. 5,585,245 and PCT WO 98/44350.

A "vector" is capable of transferring a gene sequence to a target cell. Typically, "vector construct", "expression vector", "expression construct", "expression cassette" and "gene transfer vector" refer to any nucleic acid construct that can be induced to express a gene of interest and transfer a gene sequence to a target cell. Thus, the term includes cloning and expression vehicles as well as integrating vectors.

"Reporter gene" or "reporter sequence" refers to any sequence that produces a protein product that is easily measured, preferably, though not necessarily, in a routine assay. Suitable reporter genes include, but are not limited to, sequences encoding proteins that mediate antibiotic resistance (e.g., ampicillin resistance, neomycin resistance, G418 resistance, puromycin resistance), sequences encoding colored or fluorescent or luminescent proteins (e.g., green fluorescent protein, enhanced green fluorescent protein, red fluorescent protein, luciferase), and proteins that mediate enhanced cell growth and/or gene amplification (e.g., dihydrofolate reductase). Epitope tags include, for example, one or more copies of FLAG, His, myc, Tap, HA, or any detectable amino acid sequence. "Expression tags" include sequences encoding reporters that can be operably linked to a desired gene sequence to monitor expression of the gene of interest.

The terms "subject" and "patient" are used interchangeably and refer to mammals, such as human patients and non-human primates, as well as laboratory animals, such as rabbits, dogs, cats, rats, mice, and other animals. Thus, as used herein, the term "subject" or "patient" refers to any mammalian patient or subject to which an expression cassette of the present invention can be administered. Subjects of the present invention include subjects having a disorder or at risk of developing a disorder.

The term "treat" or "treatment" as used herein refers to reducing the severity and/or frequency of symptoms, eliminating symptoms and/or their underlying causes, preventing the appearance of symptoms and/or their underlying causes, and ameliorating or curing damage. Cancer and graft-versus-host disease are non-limiting examples of conditions that can be treated using the compositions and methods described herein. "Treat" and "treatment" therefore include:
(i) preventing the appearance of a disease or condition in a mammal, particularly where such mammal is predisposed to the condition but has not yet been diagnosed as having it;
(ii) inhibiting the disease or condition, i.e., arresting its onset;
(iii) alleviating the disease or condition, i.e., causing the regression of the disease or condition; or (iv) alleviating the symptoms resulting from the disease or condition, i.e., alleviating pain without addressing the underlying disease or condition.

As used herein, the terms "disease" and "condition" may be used interchangeably or may differ in that a particular disease or condition does not have a known causative agent (so the etiology has not yet been elucidated) and therefore is not yet recognized as a disease, but only as an undesirable condition or syndrome, with some specific set of symptoms identified by clinicians.

"Pharmaceutical composition" refers to a formulation of a compound of the invention and a vehicle generally accepted in the art for the delivery of a biologically active compound to a mammal, e.g., a human. Such vehicles therefore include any pharma- ceutically acceptable carrier, diluent, or excipient.

An "effective amount" or "therapeutically effective amount" refers to an amount of a compound of the present invention that, when administered to a mammal, preferably a human, is sufficient to effect treatment in the mammal, preferably a human. The amount of a composition of the present invention that constitutes a "therapeutically effective amount" will vary depending on the compound, the condition and its severity, the mode of administration, and the age of the mammal being treated, but can be routinely determined by one of skill in the art having regard to their own knowledge and this disclosure.

DNA-binding domains Described herein are compositions comprising DNA-binding domains that specifically bind to target sites in any gene, including HLA genes or HLA regulators (including the B2M gene). Any DNA-binding domain can be used in the compositions and methods disclosed herein, including but not limited to zinc finger DNA-binding domains, TALE DNA-binding domains, DNA-binding portions of CRISPR/Cas nucleases (sgRNAs) or meganuclease-derived DNA-binding domains. The DNA-binding domain can bind to any target sequence within a gene, including but not limited to target sequences of 12 or more nucleotides set forth in any of SEQ ID NOs: 6-48.

In certain embodiments, the DNA binding domain comprises a zinc finger protein. Preferably, the zinc finger protein is non-naturally occurring in that it has been engineered to bind to a selected target site. See, e.g., Beerli et al. (2002) Nature Biotechnol. 20:135-141; Pabo et al. (2001) Ann. Rev. Biochem. 70:313-340; Isalan et al. (2001) Nature Biotechnol. 19:656-660; Segal et al. (2001) Curr. Opin. Biotechnol. 12:632-637; Choo et al. (2000) Curr. Opin. Struct. See Biol. 10:411-416; U.S. Patent Nos. 6,453,242; 6,534,261; 6,599,692; 6,503,717; 6,689,558; 7,030,215; 6,794,136; 7,067,317; 7,262,054; 7,070,934; 7,361,635; 7,253,273; and U.S. Patent Application Publication Nos. 2005/0064474; 2007/0218528; and 2005/0267061, all of which are incorporated by reference in their entireties. In certain embodiments, the DNA binding domain comprises a zinc finger protein as disclosed in U.S. Patent Application Publication No. 2012/0060230, the entirety of which is incorporated herein by reference (e.g., Table 1).

Engineered zinc finger binding domains can have novel binding specificities compared to naturally occurring zinc finger proteins. Engineering methods include, but are not limited to, rational design and various types of selection. Rational design includes, for example, using databases containing triplet (or tetrad) nucleotide sequences and individual zinc finger amino acid sequences, where each triplet or tetrad nucleotide sequence is associated with one or more amino acid sequences of zinc fingers that bind to a particular triplet or tetrad sequence. See, for example, U.S. Patent Nos. 6,453,242 and 6,534,261, which are incorporated by reference in their entireties.

Exemplary selection methods, including phage display and two-hybrid systems, are disclosed in U.S. Patent Nos. 5,789,538; 5,925,523; 6,007,988; 6,013,453; 6,410,248; 6,140,466; 6,200,759; and 6,242,568; as well as WO 98/37186; WO 98/53057; WO 00/27878; WO 01/88197 and GB 2,338,237. Additionally, enhanced binding specificity of zinc finger binding domains is described, for example, in U.S. Patent No. 6,794,136.

Further, as disclosed in these and other references, zinc finger domains and/or multi-finger zinc finger proteins can be linked together using any suitable linker sequence, including, for example, linkers of 5 or more amino acids in length. See also U.S. Pat. Nos. 6,479,626; 6,903,185; and 7,153,949 for exemplary linker sequences of 6 or more amino acids in length. The proteins described herein can include any combination of suitable linkers between the individual zinc fingers of the protein. Furthermore, enhanced binding specificity for zinc finger binding domains is described, for example, in U.S. Pat. No. 6,794,136.

Methods for targeting sites; selection of ZFPs and designing and constructing fusion proteins (and the polynucleotides encoding same) are known to those of skill in the art and are described in U.S. Patent Nos. 6,140,081, 5,789,538, 6,453,242, 6,534,261, 5,925,523, 6,007,988, 6,013,453, 6,200,7 59, WO95/19431, WO96/06166, WO98/53057, WO98/54311, WO00/27878, WO01/60970, WO01/88197, WO02/099084, WO98/53058, WO98/53059, WO98/53060, WO02/016536 and WO03/016496.

Further, as disclosed in these and other references, zinc finger domains and/or multi-fingered zinc finger proteins can be linked together using any suitable linker sequence, including, for example, linkers of 5 or more amino acids in length. See also U.S. Pat. Nos. 6,479,626, 6,903,185, and 7,153,949 for exemplary linker sequences of 6 or more amino acids in length. The proteins described herein can include any combination of suitable linkers between the individual zinc fingers of the protein.

In certain embodiments, the DNA binding domain is an engineered zinc finger protein that binds (in a sequence-specific manner) to a target site in the B2M gene or a B2M-regulated gene and modulates expression of the B2M gene. In some embodiments, the zinc finger protein binds to a target site in B2M, while in other embodiments, the zinc finger binds to a target site in B2M.

Usually, ZFPs contain at least three fingers. Certain ZFPs contain four, five or six fingers. ZFPs containing three fingers typically recognize target sites containing 9 or 10 nucleotides; ZFPs containing four fingers typically recognize target sites containing 12 to 14 nucleotides; whereas ZFPs with six fingers can recognize target sites containing 18 to 21 nucleotides. ZFPs may be fusion proteins that contain one or more regulatory domains, which may be transcriptional activation or repression domains.

In some embodiments, the DNA binding domain may be derived from a nuclease. For example, the recognition sequences of homing endonucleases and meganucleases (e.g., I-SceI, I-CeuI, PI-PspI, PI-Sce, I-SceIV, I-CsmI, I-PanI, I-SceII, I-PpoI, I-SceIII, I-CreI, I-TevI, I-TevII, and I-TevIII) are known. See also U.S. Patent No. 5,420,032; U.S. Patent No. 6,833,252; Belfort et al. (1997) Nucleic Acids Res. 25:3379-3388; Dujon et al. (1989) Gene 82:115-118; Perler et al. (1994) Nucleic Acids Res. 22, 1125-1127; Jasin (1996) Trends Genet. 12:224-228; Gimble et al. (1996) J. Mol. Biol. 263:163-180; Argast et al. (1998) J. Mol. Biol. 280:345-353 and the New England Biolabs catalogue. Additionally, the DNA binding specificity of homing endonucleases and meganucleases can be engineered to bind to non-natural target sites. See, e.g., Chevalier et al. (2002) Molec. Cell 10:895-905; Epinat et al. (2003) Nucleic Acids Res. 31:2952-2962; Ashworth et al. (2006) Nature 441:656-659; Paques et al. (2007) Current Gene Therapy 7:49-66; U.S. Patent Application Publication No. 20070117128.

In other embodiments, the DNA binding domain comprises an engineered domain from a TAL effector similar to those from the plant pathogens Xanthomonas (see Boch et al., (2009) Science 326:1509-1512 and Moscow and Bogdanove, (2009) Science 326:1501) and Ralstonia (see Heuer et al., (2007) Applied and Environmental Microbiology 73(13):4379-4384); U.S. Patent Application Nos. 20110301073 and 20110145940. Plant pathogenic bacteria of the genus Xanthomonas are known to cause many diseases in important crop plants. Xanthomonas virulence depends on a conserved type III secretion (T3S) system that injects over 25 different effector proteins into plant cells. Transcription activator-like effectors (TALEs), which mimic plant transcription activators to manipulate the plant transcriptome, are among these injected proteins (see Kay et al. (2007) Science 318:648-651). These proteins contain DNA-binding and transcription activation domains. One of the best-characterized TALEs is AvrBs3 from Xanthomonas campestgris pv. Vesicatoria (see Bonas et al. (1989) Mol Gen Genet 218:127-136 and WO2010079430). TALEs contain a centralized domain of tandem repeats, each containing approximately 34 amino acids, which are important for the DNA binding specificity of these proteins. In addition, they contain nuclear localization sequences and acidic transcription activation domains (for review, see Schornack S et al. (2006) J Plant Physiol 163(3):256-272). Furthermore, in the plant pathogenic bacterium Ralstonia solanacearum, R. Two genes, called brg11 and hpx17, were found in C. solanacearum biovar 1 strain GMI1000 and biovar 4 strain RS1000 that are homologous to the Xanthomonas AvrBs3 family (see Heuer et al. (2007) Appl and Envir Micro 73(13):4379-4384). These genes are 98.9% identical to each other in nucleotide sequence, but differ by a 1,575 bp deletion in the repeat domain of hpx17. However, both gene products share less than 40% sequence identity with Xanthomonas AvrBs3 family proteins.

The specificity of these TAL effectors depends on the sequence found in the tandem repeats. The repeats contain approximately 102 base pairs, and the repeats are typically 91-100% homologous to each other (Bonas et al., supra). The polymorphisms in the repeats are usually located at positions 12 and 13, and there appears to be a one-to-one correspondence between the identity of the hypervariable diresidues (repeat variable diresidues or RVD regions) at positions 12 and 13 and the identity of adjacent nucleotides in the target sequence of the TAL effector (see Moscou and Bogdanove, (2009) Science 326:1501 and Boch et al. (2009) Science 326:1509-1512). Experimentally, the natural codes for DNA recognition of these TAL effectors were determined such that the HD sequence (repeat variable di-residue or RVD region) at positions 12 and 13 results in binding to cytosine (C), NG binds to T, NI binds to A, C, G or T, NN binds to A or G, and ING binds to T. These DNA-binding repeats were assembled into proteins with new combinations and multiple repeats to create artificial transcription factors that can interact with new sequences to activate the expression of non-endogenous reporter genes in plant cells (Boch et al., supra). Engineered TAL proteins were linked to a FokI cleavage half-domain to obtain TAL effector domain nuclease fusions (TALENs), including TALENs with atypical RVDs. See, e.g., U.S. Pat. No. 8,586,526.

In some embodiments, the TALEN comprises an endonuclease (e.g., FokI) cleavage domain or cleavage half-domain. In other embodiments, the TALE nuclease is a megaTAL. These megaTAL nucleases are fusion proteins that comprise a TALE DNA binding domain and a meganuclease cleavage domain. The meganuclease cleavage domain is active as a monomer and does not require dimerization for activity (see Boissel et al., (2013) Nucl Acid Res: 1-13, doi: 10.1093/nar/gkt1224).

In yet further embodiments, the nuclease comprises a compact TALEN. These are single-stranded fusion proteins linking a TALE DNA-binding domain to a TevI nuclease domain. Depending on where the TALE DNA-binding domain is located relative to the TevI nuclease domain, the fusion protein can act as a nickase localized by the TALE region or can cause a double-stranded break (see Beurdeley et al. (2013) Nat Comm: pp. 1-8, DOI: 10.1038/ncomms2782). Additionally, the nuclease domain can also exhibit DNA-binding functionality. Any TALEN can be used in combination with additional TALENs, including one or more megaTALEs, such as one or more TALENs (cTALENs or FokI-TALENs).

Further, as disclosed in these and other references, zinc finger domains and/or multi-fingered zinc finger proteins or TALEs can be linked together using any suitable linker sequence, including, for example, linkers of 5 or more amino acids in length. See also U.S. Pat. Nos. 6,479,626, 6,903,185; and 7,153,949 for exemplary linker sequences of 6 or more amino acids in length. The proteins described herein can include any combination of suitable linkers between the individual zinc fingers of the protein. Furthermore, enhanced binding specificity for zinc finger binding domains is described, for example, in U.S. Pat. No. 6,794,136.

In certain embodiments, the DNA binding domain is part of a CRISPR/Cas nuclease system that includes a single guide RNA (sgRNA) that binds to DNA. See, e.g., U.S. Patent No. 8,697,359 and U.S. Patent Application Publication Nos. 20150056705 and 20150159172. The CRISPR (clustered regularly interspersed short palindromic repeats) locus, which encodes the RNA components of the system, and the protein-encoding cas (CRISPR-associated) locus (Jansen et al., 2002, Mol. Microbiol. 43:1565-1575; Makarova et al., 2002, Nucleic Acids Res. 30:482-496; Makarova et al., 2006, Biol. Direct 1:7; Haft et al., 2005, PLoS Comput. Biol. 1:e60) constitute the genetic sequence of the CRISPR/Cas nuclease system. CRISPR loci in microbial hosts contain a combination of CRISPR-associated (Cas) genes as well as non-coding RNA elements that can program the specificity of CRISPR-mediated nucleic acid cleavage.

Type II CRISPR is one of the best-characterized systems and executes targeted DNA double-strand breaks in four sequential steps. First, two non-coding RNAs, the pre-crRNA array and the tracrRNA, are transcribed from the CRISPR locus. Second, the tracrRNA hybridizes to the repeat region of the pre-crRNA and mediates processing of the pre-crRNA into mature crRNAs containing individual spacer sequences. Third, the mature crRNA:tracrRNA complex guides a functional domain (e.g., a nuclease such as Cas) to the target DNA via Watson-Crick base pairing between the spacer on the crRNA and the protospacer on the target DNA next to the protospacer adjacent motif (PAM), which is an additional requirement for target recognition. Finally, Cas9 mediates cleavage of the target DNA to cause a double-strand break within the protospacer. The activity of the CRISPR/Cas system consists of three steps: (i) insertion of alien DNA sequences into the CRISPR array to prevent future attacks in a process called "adaptation", (ii) expression of the associated proteins, and expression and processing of the array, followed by (iii) RNA-mediated interference with the alien nucleic acid. Thus, in bacterial cells, some of the so-called "Cas" proteins are involved in the natural function of the CRISPR/Cas system, playing a role in functions such as the insertion of alien DNA, etc.

In certain embodiments, the Cas protein may be a "functional derivative" of a naturally occurring Cas protein. A "functional derivative" of a native sequence polypeptide is a compound that has qualitative biological properties in common with the native sequence polypeptide. "Functional derivatives" include, but are not limited to, fragments of native sequences, as well as derivatives of native sequence polypeptides and fragments thereof, provided that they have biological activity in common with the corresponding native sequence polypeptide. The biological activity contemplated herein is the ability of the functional derivative to hydrolyze a DNA substrate into fragments. The term "derivative" encompasses amino acid sequence variants of the polypeptide, covalent modifications, and fusions thereof, such as derivative Cas proteins. Suitable derivatives of Cas polypeptides or fragments thereof include, but are not limited to, mutants, fusions, covalent modifications, or fragments thereof, of Cas proteins. Cas proteins, including Cas proteins or fragments thereof, as well as derivatives of Cas proteins or fragments thereof, can be obtained from cells or can be chemically synthesized, or obtained by a combination of these two procedures. The cells may be cells that naturally produce the Cas protein, or cells that naturally produce the Cas protein and have been engineered to produce an endogenous Cas protein at higher expression levels, or to produce the Cas protein from an exogenously introduced nucleic acid that encodes the same or a different Cas than the endogenous Cas. In some cases, the cells do not naturally produce the Cas protein and are engineered to produce the Cas protein. In some embodiments, the Cas protein is a small Cas9 ortholog for delivery via an AAV vector (Ran et al. (2015) Nature 510:186).

In some embodiments, the DNA binding domain is part of the TtAgo system (see Swarts et al., supra; Sheng et al., supra). In eukaryotes, gene silencing is mediated by the Argonaute (Ago) family of proteins. In this paradigm, Ago binds small (19-31 nt) RNAs. This protein-RNA silencing complex recognizes the target RNA through Watson-Crick base pairing between the small RNA and the target and endonucleolytically cleaves the target RNA (Vogel (2014) Science 344:972-973). In contrast, prokaryotic Ago proteins bind small single-stranded DNA fragments and likely function to detect and remove foreign (often viral) DNA (Yuan et al., (2005) Mol. Cell 19:405; Olovnikov et al., (2013) Mol. Cell 51:594; Swarts et al., supra). Exemplary prokaryotic Ago proteins include those from Aquifex aeolicus, Rhodobacter sphaeroides, and Thermus thermophilus.

One of the best-characterized prokaryotic Ago proteins is from T. thermophilus (TtAgo; Swarts et al., supra). TtAgo associates with 15 nt or 13-25 nt single-stranded DNA fragments at the 5' phosphate group. This "guide DNA" that binds to TtAgo serves to guide the protein-DNA complex to bind to Watson-Crick complementary DNA sequences in a third molecule of DNA. Once the sequence information in these guide DNAs allows the target DNA to be identified, the TtAgo-guide DNA complex cleaves the target DNA. Such a mechanism is also supported by the structure of the TtAgo-guide DNA complex while bound to its target DNA (G. Sheng et al., supra). Ago from Rhodobacter sphaeroides (RsAgo) has similar properties (Olivnikov et al., supra).

Exogenous guide DNA of any DNA sequence can be loaded into the TtAgo protein (Swarts et al., supra). Because the specificity of TtAgo cleavage is directed by the guide DNA, a TtAgo-DNA complex formed with an exogenous, investigator-specified guide DNA will thus direct TtAgo target DNA cleavage to the complementary investigator-specified target DNA. In this manner, targeted double-stranded breaks can be generated in DNA. The use of a TtAgo-guide DNA system (or an orthologous Ago-guide DNA system from another organism) allows for targeted cleavage of genomic DNA within a cell. Such breaks can be single-stranded or double-stranded. For cleavage of mammalian genomic DNA, it would be preferable to use a version of the TtAgo codon that is optimized for expression in mammalian cells. Additionally, it may be preferable to treat cells with a TtAgo-DNA complex formed in vitro, where the TtAgo protein is fused to a cell-penetrating peptide. Additionally, it may be preferable to use a version of the TtAgo protein that has been altered through mutagenesis to have improved activity at 37° C. Ago-RNA mediated DNA cleavage could be used to affect a panopoly of outcomes including gene knockout, targeted gene addition, gene correction, targeted gene deletion using techniques standard in the art for leveraging DNA cleavage.

Therefore, any DNA binding domain can be used.

Fusion Molecules Fusion molecules comprising a DNA binding domain described herein (e.g., a CRISPR/Cas component such as a ZFP or TALE, single guide RNA, etc.) in association with a heterologous regulatory (functional) domain (or a functional fragment thereof) are also provided. Common domains include, for example, transcription factor domains (activators, repressors, coactivators, corepressors), silencers, oncogenes (e.g., myc, jun, fos, myb, max, mad, rel, ets, bcl, myb, mos family members, etc.); DNA repair enzymes and their associated factors and modifiers; DNA reorganization enzymes and their associated factors and modifiers; chromatin-associated proteins and their modifiers (e.g., kinases, acetylases, and deacetylases); and DNA modifying enzymes (e.g., methyltransferases, topoisomerases, helicases, ligases, kinases, phosphatases, polymerases, endonucleases) and their associated factors and modifiers. Such fusion molecules include transcription factors and transcriptional regulatory domains containing the DNA-binding domains described herein, as well as nucleases containing the DNA-binding domains and one or more nuclease domains.

Suitable domains for achieving activation (transcription activation domains) include the HSV VP16 activation domain (see, e.g., Hagmann et al., J. Virol. 71:5952-5962 (1997)), nuclear hormone receptors (see, e.g., Torchia et al., Curr. Opin. Cell. Biol. 10:373-383 (1998)); the p65 subunit of nuclear factor kappa B (Bitko & Barik, J. Virol. 72:5610-5618 (1998) and Doyle & Hunt, Neuroreport 8:2937-2942 (1997)), Liu et al., Cancer Gene Ther. 5:3-28 (1998)), or artificial chimeric functional domains such as VP64 (Beerli et al., (1998) Proc. Natl. Acad. Sci. USA 95:14623-33), and degrons (Molinari et al. (1999) EMBO J. 18:6439-6447). Further exemplary activation domains include Oct 1, Oct-2A, Sp1, AP-2, and CTF1 (Seipel et al., EMBO J. 11:4961-4968 (1992) as well as p300, CBP, PCAF, SRC1 PvALF, AtHD2A, and ERF-2. See, e.g., Robyr et al. (2000) Mol. Endocrinol. 14:329-347; Collingwood et al. (1999) J. Mol. Endocrinol. 23:255-275; Leo et al. (2000) Gene 245:1-11; Manteuffel-Cymborowska (1999) Acta Biochim. Pol. 46:77-89, McKenna et al. (1999) J. Steroid Biochem. Mol. Biol. 69:3-12, Malik et al. (2000) Trends Biochem. Sci. 25:277-283, and Lemon et al. (1999) Curr. Opin. Genet. Dev. 9:499-504. Further exemplary activation domains include, but are not limited to, OsGAI, HALF-1, C1, AP1, ARF-5, -6, -7 and -8, CPRF1, CPRF4, MYC-RP/GP, and TRAB1. See, e.g., Ogawa et al. (2000) Gene 245:21-29, Okanami et al. (1996) Genes Cells 245:21-29, and Ogawa et al. (1996) Genes Cells 245:21-29. 1:87-99, Goff et al. (1991) Genes Dev. 5:298-309, Cho et al. (1999) Plant Mol. Biol. 40:419-429, Ulmason et al. (1999) Proc. Natl. Acad. Sci. USA 96:5844-5849, Sprenger-Haussels et al. (2000) Plant J. 22:1-8, Gong et al. (1999) Plant Mol. Biol. 41:33-44, and Hobo et al. (1999) Proc. Natl. Acad. Sci. USA 96:15,348-15,353.

It will be apparent to one skilled in the art that in forming a fusion protein (or a nucleic acid encoding the same) between a DNA binding domain and a functional domain, either an activation domain or a molecule that interacts with an activation domain is suitable as the functional domain. Essentially, any molecule capable of recruiting an activation complex and/or an activation activity (e.g., histone acetylation, etc.) to a target gene is useful as an activation domain of a fusion protein. Chromatin remodeling proteins and/or methyl-binding domain proteins, such as insulator domains, localization domains, and ISWI-containing domains, suitable for use as functional domains in fusion molecules are described, for example, in U.S. Pat. No. 7,053,264.

Exemplary repression domains include, but are not limited to, KRAB A/B, KOX, TGF beta-inducible early gene (TIEG), v-erbA, SID, MBD2, MBD3, members of the DNMT family (e.g., DNMT1, DNMT3A, DNMT3B), Rb, and MeCP2. See, e.g., Bird et al. (1999) Cell 99:451-454, Tyler et al. (1999) Cell 99:443-446, Knoepfler et al. (1999) Cell 99:447-450, and Robertson et al. (2000) Nature Genet. 25:338-342. Further exemplary repression domains include, but are not limited to, ROM2 and AtHD2A. See, e.g., Chem et al. (1996) Plant Cell 8:305-321 and Wu et al. (2000) Plant J. 22:19-27.

Fusion molecules are constructed by methods of cloning and biochemical conjugation well known to those of skill in the art. Fusion molecules include a DNA-binding domain (e.g., ZFP, TALE, sgRNA) and a functional domain (e.g., a transcriptional activation or repression domain). Fusion molecules also optionally include a nuclear localization signal (e.g., from SV40 medium T-antigen) and an epitope tag (e.g., FLAG and hemagglutinin). Fusion proteins (and the nucleic acids encoding them) are designed such that the translational reading frame is preserved within the components of the fusion.

Fusions between polypeptide components of a functional domain (or functional fragment thereof) on the one hand and a non-protein DNA binding domain (e.g., antibiotics, interfering substances, minor groove binders, nucleic acids) on the other hand are constructed by methods of biochemical conjugation known to those of skill in the art. See, e.g., the Pierce Chemical Company (Rockford, IL) Catalogue. Methods and compositions for making fusions between minor groove binders and polypeptides have been described. Mapp et al. (2000) Proc. Natl. Acad. Sci. USA 97:3930-3935. Furthermore, the single guide RNA of the CRISPR/Cas system associates with the functional domain to form an active transcription factor and nuclease.

In certain embodiments, the target site is present in an accessible region of cellular chromatin. The accessible region can be determined, for example, as described in U.S. Patent Nos. 7,217,509 and 7,923,542. If the target site is not present in an accessible region of cellular chromatin, one or more accessible regions can be generated, for example, as described in U.S. Patent Nos. 7,785,792 and 8,071,370. In further embodiments, the DNA binding domain of the fusion molecule can bind to cellular chromatin regardless of whether its target site is in an accessible region. For example, such a DNA binding domain can bind to linker DNA and/or nucleosomal DNA. Examples of this type of "pioneer" DNA-binding domain are found in certain steroid receptors and in hepatocyte nuclear factor 3 (HNF3) (Cordingley et al. (1987) Cell 48:261-270; Pina et al. (1990) Cell 60:719-731; and Cirillo et al. (1998) EMBO J. 17:244-254).

The fusion molecules can be formulated with a pharma- ceutically acceptable carrier as known to those of skill in the art. See, e.g., Remington's Pharmaceutical Sciences, 17th ed., 1985, and U.S. Pat. Nos. 6,453,242 and 6,534,261.

The functional component/domain of the fusion molecule can be selected from any of a variety of different components that are capable of affecting transcription of a gene when the fusion molecule binds to a target sequence via its DNA binding domain. Functional components thereby include, but are not limited to, various transcription factor domains such as activators, repressors, coactivators, corepressors and silencers.

Further exemplary functional domains are disclosed, for example, in U.S. Patent Nos. 6,534,261 and 6,933,113.

Functional domains can also be selected that are controlled by exogenous small molecules or ligands. For example, RheoSwitch® technology can be used, where a functional domain only assumes its active conformation in the presence of an external RheoChem™ ligand (see, e.g., US20090136465). Thus, a ZFP can be operably linked to a regulatable functional domain, and the activity of the resulting ZFP-TF is regulated by an external ligand.

Nucleases In certain embodiments, the fusion molecule comprises a DNA binding domain associated with a cleavage (nuclease) domain. Such genetic modification can be achieved using nucleases, e.g., engineered nucleases. Engineered nuclease technology is based on the engineering of naturally occurring DNA binding proteins. For example, the engineering of homing endonucleases with tailored DNA binding specificity has been described. Chames et al. (2005) Nucleic Acids Res 33(20):e178; Arnould et al. (2006) J. Mol. Biol. 355:443-458. In addition, the engineering of ZFPs has also been described. See, e.g., U.S. Patent Nos. 6,534,261, 6,607,882, 6,824,978, 6,979,539, 6,933,113, 7,163,824, and 7,013,219.

Furthermore, ZFPs and/or TALEs can be fused to nuclease domains to create ZFNs and TALENs - functional entities that can recognize their intended nucleic acid target through their engineered (ZFP or TALE) DNA binding domain and cause DNA cleavage near the DNA binding site through nuclease activity.

Thus, the methods and compositions described herein are broadly applicable and can include any nuclease of interest. Non-limiting examples of nucleases include meganucleases, TALENs, and zinc finger nucleases. Nucleases can include heterologous DNA binding and cleavage domains (e.g., zinc finger nucleases, meganuclease DNA binding domains that include heterologous cleavage domains), or alternatively, the DNA binding domain of a naturally occurring nuclease can be altered to bind to a selected target site (e.g., meganucleases engineered to bind to a site different from the cognate binding site).

In any of the nucleases described herein, the nuclease can include an engineered TALE DNA binding domain, also referred to as a TALEN, and a nuclease domain (e.g., an endonuclease and/or a meganuclease domain). Methods and compositions for engineering these TALEN proteins for robust, site-specific interaction with a user-selected target sequence have been published (see U.S. Pat. No. 8,586,526). In some embodiments, the TALEN includes an endonuclease (e.g., FokI) cleavage domain or cleavage half-domain. In other embodiments, the TALE nuclease is a megaTAL. These megaTAL nucleases are fusion proteins that include a TALE DNA binding domain and a meganuclease cleavage domain. Meganuclease cleavage domains are active as monomers and do not require dimerization for activity (see Boissel et al., (2013) Nucl Acid Res: 1-13, doi: 10.1093/nar/gkt1224). Additionally, nuclease domains can also exhibit DNA binding functionality.

In yet further embodiments, the nuclease comprises a compact TALEN (cTALEN). These are single-stranded fusion proteins that link a TALE DNA-binding domain to a TevI nuclease domain. The fusion protein can act as a nickase localized by the TALE region or can cause a double-stranded break, depending on where the TALE DNA-binding domain is located relative to the TevI nuclease domain (see Beurdeley et al. (2013) Nat Comm: pp. 1-8, DOI: 10.1038/ncomms2782). Any TALEN can be used in combination with additional TALENs (e.g., one or more TALENs with one or more megaTALs (cTALENs or FokI-TALENs)) or other DNA cleavage enzymes.

In certain embodiments, the nuclease comprises a meganuclease (homing endonuclease) or a portion thereof that exhibits cleavage activity. Naturally occurring meganucleases recognize cleavage sites of 15-40 base pairs and are generally classified into four families: the LAGLIDADG family, the GIY-YIG family, the His-Cyst box family, and the HNH family. Exemplary homing endonucleases include I-SceI, I-CeuI, PI-PspI, PI-Sce, I-SceIV, I-CsmI, I-PanI, I-SceII, I-PpoI, I-SceIII, I-CreI, I-TevI, I-TevII, and I-TevIII. Their recognition sequences are known. See also U.S. Patent No. 5,420,032; U.S. Patent No. 6,833,252; Belfort et al. (1997) Nucleic Acids Res. 25:3379-3388; Dujon et al. (1989) Gene 82:115-118; Perler et al. (1994) Nucleic Acids Res. 22, 1125-1127; Jasin (1996) Trends Genet. 12:224-228; Gimble et al. (1996) J. Mol. Biol. 263:163-180; Argast et al. (1998) J. Mol. Biol. 280:345-353 and the New England Biolabs catalog.

DNA-binding domains from naturally occurring meganucleases, primarily from the LAGLIDADG family, have been used to facilitate site-specific genome modification in plants, yeast, Drosophila, mammalian cells and mice, but this approach relies on modifications of either homologous genes that preserve the meganuclease recognition sequence (Monet et al. (1999) Biochem. Biophysics. Res. Common. 255:88-93) or modifications to the genome prior to engineering that introduce the recognition sequence (Route et al. (1994) Mol. Cell. Biol. 14:8096-106; Chilton et al. (2003) Plant Physiology. 133:956-65; Puchta et al. (1996) Proc. Natl. Acad. Sci. USA 93:5055-60; Rong et al. (2002) Genes Dev. 16:1568-81; Gouble et al. (2006), J. Gene Med. 8(5):616-622). Therefore, attempts have been made to engineer meganucleases that display novel binding specificities at medically or biotechnologically relevant sites (Porteus et al. (2005) Nat. Biotechnol. 23:967-73; Sussman et al. (2004) J. Mol. Biol. 342:31-41; Epinat et al. (2003) Nucleic Acids Res. 31:2952-62; Chevalier et al. (2002) Molec. Cell 10:895-905; Epinat et al. (2003) Nucleic Acids Res. 31:2952-2962; Ashworth et al. (2006) Nature 441:656-659; Paques et al. (2007) Current Gene Therapy 14:111-112; 7:49-66, U.S. Patent Application Publication Nos. 20070117128, 20060206949, 20060153826, 20060078552, and 20040002092. Additionally, a naturally occurring or engineered DNA binding domain from a meganuclease can be operably linked to a cleavage domain from a heterologous nuclease (e.g., FokI), and/or a cleavage domain from a meganuclease can be operably linked to a heterologous DNA binding domain (e.g., ZFP or TALE).

In other embodiments, the nuclease is a zinc finger nuclease (ZFN) or a TALE DNA binding domain nuclease fusion (TALEN). ZFNs and TALENs contain a DNA binding domain (a zinc finger protein or a TALE DNA binding domain) and a cleavage domain or cleavage half-domain (e.g., from a restriction and/or meganuclease described herein) engineered to bind to a target site in a gene of choice.

As detailed above, zinc finger binding domains and TALE DNA binding domains can be engineered to bind to a sequence of choice. See, e.g., Beerli et al. (2002) Nature Biotechnol. 20:135-141; Pabo et al. (2001) Ann. Rev. Biochem. 70:313-340; Isalan et al. (2001) Nature Biotechnol. 19:656-660; Segal et al. (2001) Curr. Opin. Biotechnol. 12:632-637; Choo et al. (2000) Curr. Opin. Struct. Biol. 10:411-416. Engineered zinc finger binding domains or TALE proteins can have novel binding specificities compared to naturally occurring proteins. Engineering methods include, but are not limited to, rational design and various types of selection. Rational design includes, for example, using a database containing triplet (or quadruplet) nucleotide sequences and individual zinc finger or TALE amino acid sequences, where each triplet or quadruplet nucleotide sequence is associated with one or more amino acid sequences of zinc finger or TALE repeat units that bind to the particular triplet or quadruplet sequence. See, for example, U.S. Patent Nos. 6,453,242 and 6,534,261, which are incorporated herein by reference in their entireties.

Methods for selecting target sites and for designing and constructing fusion proteins (and the polynucleotides encoding same) are known to those of skill in the art and are described in detail in U.S. Pat. Nos. 7,888,121 and 8,409,861, which are incorporated by reference in their entireties.

Further, as disclosed in these and other references, zinc finger domains, TALEs and/or multi-fingered zinc finger proteins can be linked together using any suitable linker sequence, including, for example, linkers of 5 or more amino acids in length. For exemplary linker sequences of 6 or more amino acids in length, see, for example, U.S. Pat. Nos. 6,479,626; 6,903,185; and 7,153,949. The proteins described herein can include any combination of suitable linkers between the individual zinc fingers of the protein. See also U.S. Pat. No. 8,772,453.

Nucleases such as ZFNs, TALENs and/or meganucleases can thereby comprise any DNA binding domain and any nuclease (cleavage) domain (cleavage domain, cleavage half-domain). As noted above, the cleavage domain may be heterologous to the DNA binding domain, for example a zinc finger or TAL effector DNA binding domain and a cleavage domain from a nuclease, or a cleavage domain from a nuclease different from the meganuclease DNA binding domain. Heterologous cleavage domains can be obtained from any endonuclease or exonuclease. Exemplary endonucleases from which a cleavage domain can be derived include, but are not limited to, restriction endonucleases and homing endonucleases. See, for example, Catalog 2002-2003, New England Biolabs, Beverly, MA; and Belfort et al. (1997) Nucleic Acids Res. 25:3379-3388. Additional enzymes that cleave DNA are known (e.g., S1 nuclease; mung bean nuclease; pancreatic DNase I; micrococcal nuclease; yeast HO endonuclease; see also Linn et al. (eds.) Nucleases, Cold Spring Harbor Laboratory Press, 1993). One or more of these enzymes (or functional fragments thereof) can be used as a source of cleavage domains and cleavage half-domains.

Similarly, as described above, the cleavage half-domains that require dimerization for cleavage activity can be derived from any nuclease or portion thereof. Generally, if two fusion proteins contain cleavage half-domains, then two fusion proteins are required for cleavage. Alternatively, a single protein containing two cleavage half-domains can be used. The two cleavage half-domains can be derived from the same endonuclease (or functional fragments thereof), or each cleavage half-domain can be derived from a different endonuclease (or functional fragments thereof). Furthermore, the target sites for the two fusion proteins are preferably positioned with respect to each other such that binding of the two fusion proteins to their respective target sites places the cleavage half-domains in a spatial orientation with each other that allows the cleavage half-domains to form a functional cleavage domain, e.g., by dimerization. Thus, in certain embodiments, the proximal ends of the target sites are separated by 5-8 nucleotides or 15-18 nucleotides. However, any integer number of nucleotides or nucleotide pairs can intervene between the two target sites (e.g., 2 to 50 nucleotide pairs, or more). Generally, the cleavage site is between the target sites, but can be one or more kilobases away from the cleavage site, including between 1-50 base pairs (or any value in between, including 1-5, 1-10, and 1-20 base pairs), between 1-100 base pairs (or any value in between), between 100-500 base pairs (or any value in between), between 500-1000 base pairs (or any value in between), or even more than 1 kb.

Restriction endonucleases (restriction enzymes) exist in many species and are capable of sequence-specific binding to DNA (at a recognition site) and cleaving the DNA at or near the binding site. Certain restriction enzymes (e.g., type IIS) cleave DNA at a site distant from the recognition site and have separable binding and cleavage domains. For example, the type IIS enzyme Fok I catalyzes a double-stranded cleavage of DNA 9 nucleotides from its recognition site on one strand and 13 nucleotides from its recognition site on the other strand. See, e.g., U.S. Patent Nos. 5,356,802; 5,436,150 and 5,487,994; and Li et al. (1992) Proc. Natl. Acad. Sci. USA 89:4275-4279; Li et al. (1993) Proc. Natl. Acad. Sci. USA 90:2764-2768; Kim et al. (1994a) Proc. Natl. Acad. Sci. USA 91:883-887; Kim et al. (1994b) J. Biol. Chem. 269:31,978-31,982. Thus, in one embodiment, the fusion protein comprises a cleavage domain (or cleavage half-domain) from at least one Type IIS restriction enzyme, which may or may not be engineered, and one or more zinc finger binding domains.

An exemplary type IIS restriction enzyme whose cleavage domain can be separated from the binding domain is Fok I. This particular enzyme is active as a dimer. Bitinaite et al. (1998) Proc. Natl. Acad. Sci. USA 95:10,570-10,575. Thus, for purposes of this disclosure, the portion of the Fok I enzyme used in the disclosed fusion proteins is considered a cleavage half-domain. Thus, for targeted double-stranded cleavage and/or targeted replacement of cellular sequences using zinc finger-Fok I fusions, two fusion proteins, each containing a Fok I cleavage half-domain, can be used to reconstitute a catalytically active cleavage domain. Alternatively, a single polypeptide molecule containing a zinc finger binding domain and two Fok I cleavage half-domains can be used. Parameters for targeted cleavage and targeted sequence alteration using zinc finger-Fok I fusions are provided elsewhere in this disclosure.

A cleavage domain or cleavage half-domain can be any portion of a protein that retains cleavage activity or the ability to multimerize (e.g., dimerize) to form a functional cleavage domain.

Exemplary Type IIS restriction enzymes are described in International Publication WO 07/014275, which is incorporated herein in its entirety. Additional restriction enzymes also contain separable binding and cleavage domains and are contemplated by this disclosure. See, e.g., Roberts et al. (2003) Nucleic Acids Res. 31:418-420.

In certain embodiments, the cleavage domain comprises one or more engineered cleavage half-domains (also referred to as dimerization domain mutants) that minimize or prevent homodimerization, as described, for example, in U.S. Pat. Nos. 7,914,796; 8,034,598 and 8,623,618 and U.S. Patent Publication No. 20110201055, the entire disclosures of which are incorporated herein by reference in their entireties. Amino acid residues at positions 446, 447, 479, 483, 484, 486, 487, 490, 491, 496, 498, 499, 500, 531, 534, 537 and 538 of Fok I are all targets for affecting dimerization of the Fok I cleavage half-domain.

Exemplary engineered cleavage half-domains of Fok I that form obligate heterodimers include a pair in which the first cleavage half-domain contains mutations at amino acid residues 490 and 538 of Fok I and the second cleavage half-domain contains mutations at amino acid residues 486 and 499.

Thus, in one embodiment, the mutation at 490 replaces Glu (E) with Lys (K); the mutation at 538 replaces Iso (I) with Lys (K); the mutation at 486 replaces Gln (Q) with Glu (E); and the mutation at position 499 replaces Iso (I) with Lys (K). Specifically, the engineered cleavage half-domains described herein were prepared by mutating positions 490 (E→K) and 538 (I→K) of one cleavage half-domain to produce an engineered cleavage half-domain designated "E490K:I538K" and mutating positions 486 (Q→E) and 499 (I→L) of another cleavage half-domain to produce an engineered cleavage half-domain designated "Q486E:I499L". The engineered cleavage half-domains described herein are obligate heterodimer mutants that minimize or abolish aberrant cleavage. See, for example, U.S. Patent Nos. 7,914,796 and 8,034,598, the disclosures of which are incorporated by reference in their entirety for all purposes. In certain embodiments, the engineered cleavage half-domains contain mutations at positions 486, 499, and 496 (numbered relative to wild-type FokI), such as replacing the wild-type Gln (Q) residue at position 486 with a Glu (E) residue, the wild-type Iso (I) residue at position 499 with a Leu (L) residue, and the wild-type Asn (N) residue at position 496 with an Asp (D) or Glu (E) residue (also referred to as the "ELD" domain and "ELE" domain, respectively). In other embodiments, the engineered cleavage half-domains contain mutations at positions 490, 538, and 537 (numbered relative to wild-type FokI); for example, mutations replacing the wild-type Glu (E) residue at position 490 with a Lys (K) residue, the wild-type Iso (I) residue at position 538 with a Lys (K) residue, and the wild-type His (H) residue at position 537 with a Lys (K) residue or an Arg (R) residue (also referred to as the "KKK" domain and the "KKR" domain, respectively). In other embodiments, the engineered cleavage half-domains contain mutations at positions 490 and 537 (numbered relative to wild-type FokI), e.g., replacing the wild-type Glu (E) residue at position 490 with a Lys (K) residue and the wild-type His (H) residue at position 537 with a Lys (K) or Arg (R) residue (also referred to as the "KIK" and "KIR" domains, respectively). See, e.g., U.S. Pat. Nos. 7,914,796; 8,034,598 and 8,623,618, the disclosures of which are incorporated by reference in their entireties for all purposes. In other embodiments, the engineered cleavage half-domains contain "Sharkey" and/or "Sharkey" mutations (see Guo et al. (2010) J. Mol. Biol. 400(1):96-107).

Alternatively, nucleases can be assembled in vivo at nucleic acid target sites using the so-called "split enzyme" technology (see, e.g., U.S. Patent Application Publication No. 20090068164). The components of such split enzymes can be expressed on separate expression constructs or can be linked into one open reading frame, where the individual components are separated, for example, by a self-cleaving 2A peptide or an IRES sequence. The components can be individual zinc finger binding domains or domains of meganuclease nucleic acid binding domains.

Nucleases (e.g., ZFNs and/or TALENs) can be screened for activity prior to use, e.g., in a yeast-based chromosomal system as described in U.S. Pat. No. 8,563,314.

In certain embodiments, the nuclease comprises a CRISPR/Cas system. The CRISPR (clustered regularly interspaced short palindromic repeats) locus, which encodes the RNA components of the system, and the Cas (CRISPR-associated) locus, which encodes the proteins (Jansen et al., 2002, Mol. Microbiol. 43:1565-1575; Makarova et al., 2002, Nucleic Acids Res. 30:482-496; Makarova et al., 2006, Biol. Direct 1:7; Haft et al., 2005, PLoS Comput. Biol. 1:e60) constitute the genetic sequence of the CRISPR/Cas nuclease system. CRISPR loci in microbial hosts contain a combination of CRISPR-associated (Cas) genes as well as non-coding RNA elements that can program the specificity of CRISPR-mediated nucleic acid cleavage.

Type II CRISPR is one of the best-characterized systems and executes targeted DNA double-strand breaks in four sequential steps. First, two non-coding RNAs, the pre-crRNA array and the tracrRNA, are transcribed from the CRISPR locus. Second, the tracrRNA hybridizes to the repeat region of the pre-crRNA and mediates processing of the pre-crRNA into mature crRNAs containing individual spacer sequences. Third, the mature crRNA:tracrRNA complex directs Cas9 to the target DNA through Watson-Crick base pairing between the spacer on the crRNA and the protospacer on the target DNA next to the protospacer adjacent motif (PAM), which is an additional requirement for target recognition. Finally, Cas9 mediates cleavage of the target DNA to generate a double-strand break within the protospacer. The activity of the CRISPR/Cas system consists of three steps: (i) insertion of alien DNA sequences into the CRISPR array to prevent future attacks in a process called "adaptation", (ii) expression of the associated proteins, and expression and processing of the array, followed by (iii) RNA-mediated interference with the alien nucleic acid. Thus, in bacterial cells, some of the so-called "Cas" proteins are involved in the natural function of the CRISPR/Cas system, playing a role in functions such as the insertion of alien DNA, etc.

In some embodiments, the CRISPR-Cpf1 system is used. The CRISPR-Cpf1 system identified in Francisella spp. is a class 2 CRISPR-Cas system that mediates robust DNA interference in human cells. Although functionally conserved, Cpf1 and Cas9 differ in many aspects, including their guide RNA and substrate specificity (see Fagerlund et al. (2015) Genom Bio 16:251). The main difference between Cas9 and Cpf1 proteins is that Cpf1 does not utilize tracrRNA, and therefore requires only crRNA. FnCpf1 crRNA is 42-44 nucleotides long (19 nucleotide repeats and a 23-25 nucleotide spacer) and contains a single stem-loop that allows for sequence changes that preserve secondary structure. Additionally, the Cpf1 crRNA is significantly shorter than the engineered sgRNA required by Cas9, by over 100 nucleotides, and the PAM requirement for FnCpfl is 5'-TTN-3' and 5'-CTA-3' on the displaced strand. Both Cas9 and Cpf1 create double-stranded breaks in the target DNA, but Cas9 uses its RuvC-like and HNH-like domains to create a blunt-end cut within the seed sequence of the guide RNA, while Cpf1 uses the RuvC-like domain to produce a staggered cut outside of the seed. Because Cpf1 creates a staggered cut away from the critical seed region, NHEJ does not destroy the target site, thus ensuring that Cpf1 can continue to cut the same site until the desired HDR recombination event occurs. Thus, in the methods and compositions described herein, it is understood that the term "Cas" includes both the Cas9 and Cpf1 proteins. Thus, as used herein, the "CRISPR/Cas system" refers to both the CRISPR/Cas and/or CRISPR/Cpf1 systems, including both the nuclease and/or transcription factor systems.

In certain embodiments, the Cas protein may be a "functional derivative" of a naturally occurring Cas protein. A "functional derivative" of a native sequence polypeptide is a compound that has qualitative biological properties in common with the native sequence polypeptide. "Functional derivatives" include, but are not limited to, fragments of native sequences, derivatives of native sequence polypeptides and fragments thereof, provided that they have biological activity in common with the corresponding native sequence polypeptide. The biological activity contemplated herein is the ability of the functional derivative to hydrolyze a DNA substrate into fragments. The term "derivative" encompasses amino acid sequence variants of the polypeptide, covalent modifications and fusions thereof. Suitable derivatives of Cas polypeptides or fragments thereof include, but are not limited to, mutants, fusions, covalent modifications or fragments of Cas proteins. Cas proteins, including Cas proteins or fragments thereof, as well as derivatives or fragments of Cas proteins, can be obtained from cells or chemically, or by a combination of these two procedures. The cells may be cells that naturally produce the Cas protein, or cells that naturally produce the Cas protein and have been engineered to produce an endogenous Cas protein at higher expression levels, or to produce the Cas protein from an exogenously introduced nucleic acid that encodes the same or a different Cas than the endogenous Cas. In some cases, the cells do not naturally produce the Cas protein and are engineered to produce the Cas protein.

Exemplary CRISPR/Cas nuclease systems for targeting TCR genes and other genes are disclosed, for example, in U.S. Patent Application Publication No. 20150056705.

The nuclease(s) can create one or more double-stranded and/or single-stranded breaks at the target site. In certain embodiments, the nuclease comprises a catalytically inactive cleavage domain (e.g., FokI and/or Cas protein). See, e.g., U.S. Pat. Nos. 9,200,266, 8,703,489 and Guillinger et al. (2014) Nature Biotech. 32(6):577-582. A catalytically inactive cleavage domain can act as a nickase to create a single-stranded break in combination with a catalytically active domain. Thus, two nickases can be used in combination to create a double-stranded break in a specific region. Additional nickases are known in the art, for example, McCaffery et al. (2016) Nucleic Acids Res. 44(2):e11. doi: 10.1093/nar/gkv878. Epub 2015 Oct 19.

Delivery Proteins (e.g., transcription factors, nucleases, TCR and CAR molecules), polynucleotides and/or compositions comprising the proteins and/or polynucleotides described herein can be delivered to target cells by any suitable means, including, for example, by injection of the protein and/or mRNA components.

Suitable cells include, but are not limited to, eukaryotic and prokaryotic cells and/or cell lines. Non-limiting examples of such cells or cell lines generated from such cells include T cells, COS, CHO (e.g., CHO-S, CHO-K1, CHO-DG44, CHO-DUXB11, CHO-DUKX, CHOK1SV), VERO, MDCK, WI38, V79, B14AF28-G3, BHK, HaK, NS0, SP2/0-Ag14, HeLa, HEK293 (e.g., HEK293-F, HEK293-H, HEK293-T) and perC6 cells, as well as insect cells such as Spodoptera fugiperda (Sf), or fungal cells such as Saccharomyces, Pichia and Schizosaccharomyces. In certain embodiments, the cell line is a CHO-K1, MDCK or HEK293 cell line. Suitable cells also include stem cells, such as, by way of example, embryonic stem cells, induced pluripotent stem cells (iPS cells), hematopoietic stem cells, neuronal stem cells and mesenchymal stem cells.

Methods of delivering proteins containing the DNA binding domains described herein are described, for example, in U.S. Patent Nos. 6,453,242; 6,503,717; 6,534,261; 6,599,692; 6,607,882; 6,689,558; 6,824,978; 6,933,113; 6,979,539; 7,013,219; and 7,163,824, the entire disclosures of which are incorporated herein by reference in their entireties.

The DNA-binding domains and fusion proteins comprising these DNA-binding domains described herein can also be delivered using vectors containing sequences encoding one or more DNA-binding proteins. In addition, additional nucleic acids (e.g., donors) can also be delivered via these vectors. Any vector system can be used, including, but not limited to, plasmid vectors, retroviral vectors, lentiviral vectors, adenoviral vectors, poxvirus vectors; herpesvirus vectors and adeno-associated virus vectors. See also U.S. Patent Nos. 6,534,261; 6,607,882; 6,824,978; 6,933,113; 6,979,539; 7,013,219; and 7,163,824, which are incorporated herein by reference in their entirety. Furthermore, it is clear that any of these vectors can optionally include one or more DNA-binding protein coding sequences and/or additional nucleic acids. Thus, when one or more of the DNA-binding proteins described herein, and optionally additional DNA, are introduced into a cell, they can be carried on the same vector or on different vectors. When multiple vectors are used, each vector can contain sequences encoding one or more DNA-binding proteins and the desired additional nucleic acid.

Conventional viral and non-viral based gene transfer methods can be used to introduce nucleic acids encoding engineered DNA binding proteins into cells (e.g., mammalian cells) and target tissues, as well as to co-introduce additional nucleotide sequences as desired. Such methods can be used to administer nucleic acids (e.g., encoding DNA binding proteins and/or donors) to cells in vitro. In certain embodiments, the nucleic acids are administered for in vivo or ex vivo gene therapy applications. Non-viral vector delivery systems include DNA plasmids, naked nucleic acids, and nucleic acids complexed with delivery vehicles such as liposomes, lipid nanoparticles, or poloxamers. Viral vector delivery systems include DNA and RNA viruses, which have either episomal or integrated genomes after delivery to cells. For reviews of gene therapy procedures, see Anderson, Science 256:808-813 (1992); Nabel and Felgner, TIBTECH 11:211-217 (1993); Mitani and Caskey, TIBTECH 11:162-166 (1993); Dillon, TIBTECH 11:167-175 (1993); Miller, Nature 357:455-460 (1992); Van Brunt, Biotechnology 6(10):1149-1154 (1988); Vigne, Restorative Neurology and Neuroscience 8:35-36 (1995); Kremer and Perricaudet, British Medical Bulletin 51(1):31-44 (1995); Haddada et al., Current Topics in Microbiology and Immunology, Doerfler and Boehm (eds.) (1995); and Yu et al., Gene Therapy 1:13-26 (1994).

Non-viral methods of delivery of nucleic acids include electroporation, lipofection, microinjection, biolistics, virosomes, liposomes, lipid nanoparticles, immunoliposomes, polycation or lipid:nucleic acid conjugates, naked DNA, mRNA, artificial virions and agent-enhanced DNA uptake. Sonoporation, for example using the Sonitron 2000 system (Rich-Mar), can also be used for delivery of nucleic acids. In a preferred embodiment, the nucleic acid or nucleic acids are delivered as mRNA. Also preferred is the use of capped mRNA for increased translation efficiency and/or mRNA stability. Particularly preferred is the ARCA (anti-reverse cap analog) cap or variants thereof. See U.S. Pat. Nos. 7,074,596 and 8,153,773, which are incorporated herein by reference.

Further exemplary nucleic acid delivery systems include those provided by Amaxa Biosystems (Cologne, Germany), Maxcyte, Inc. (Rockville, Maryland), BTX Molecular Delivery Systems (Holliston, Mass.) and Copernicus Therapeutics Inc. (see, e.g., US 6,008,336). Lipofection is described, for example, in US 5,049,386; US 4,946,787; and US 4,897,355, and lipofection reagents are commercially available (e.g., Transfectam™, Lipofectin™, and Lipofectamine™ RNAiMAX). Cationic and neutral lipids suitable for efficient receptor-recognition lipofection of polynucleotides include those of Felgner, WO 91/17424, WO 91/16024. Delivery can be to cells (ex vivo administration) or to target tissues (in vivo administration).

The preparation of lipid:nucleic acid complexes, including targeted liposomes such as immunolipid complexes, is well known to those of skill in the art (e.g., Crystal Science 270:404-410 (1995); Blaese et al., Cancer Gene Ther. 2:291-297 (1995); Behr et al., Bioconjugate Chem. 5:382-389 (1994); Remy et al., Bioconjugate Chem. 5:647-654 (1994); Gao et al., Gene Therapy 2:710-722 (1995); Ahmad et al., Cancer Res. 52:4817-4820 (1992); see U.S. Patent Nos. 4,186,183, 4,217,344, 4,235,871, 4,261,975, 4,485,054, 4,501,728, 4,774,085, 4,837,028 and 4,946,787).

Additional methods of delivery include the use of packaging the nucleic acid to be delivered into EnGeneIC delivery vehicles (EDVs). These EDVs are delivered specifically to target tissues using bispecific antibodies, where one arm of the bispecific antibody has specificity for the target tissue and the other for the EDV. The antibody carries the EDV to the target cell surface, where it is then carried into the cell by endocytosis. Once inside the cell, the contents are released (see MacDiarmid et al. (2009) Nature Biotechnology 27(7):643).

The use of RNA or DNA virus-based systems for delivery of nucleic acids encoding engineered DNA-binding proteins and/or donors (e.g., CAR or ACTR), if desired, takes advantage of the highly evolved processes for directing viruses to specific cells in the body and transporting the viral payload to the nucleus. Viral vectors can be administered directly to patients (in vivo) or they can be used to treat cells in vitro and the modified cells are administered to patients (ex vivo). Conventional virus-based systems for delivery of nucleic acids include, but are not limited to, retroviral, lentiviral, adenoviral, adeno-associated, vaccinia and herpes simplex virus vectors for gene transfer. Integration into the host genome is possible with retroviral, lentiviral and adeno-associated virus gene transfer methods, often resulting in long-term expression of the inserted transgene. Furthermore, high transduction efficiencies have been observed in many different cell types and target tissues.

The tropism of retroviruses can be altered by incorporating foreign envelope proteins to expand the potential target population of target cells. Lentiviral vectors are retroviral vectors that can transduce or infect non-dividing cells and generally produce high viral titers. The choice of retroviral gene transfer system depends on the target tissue. Retroviral vectors are composed of cis-acting long terminal repeats with packaging capacity for up to 6-10 kb of foreign sequences. The minimal cis-acting LTRs are sufficient for vector replication and packaging, which are then used to incorporate therapeutic genes into target cells to provide permanent transgene expression. Widely used retroviral vectors include those based on murine leukemia virus (MuLV), gibbon ape leukemia virus (GaLV), simian immunodeficiency virus (SIV), human immunodeficiency virus (HIV), and combinations thereof (see, e.g., Buchscher et al., J. Virol. 66:2731-2739 (1992); Johann et al., J. Virol. 66:1635-1640 (1992); Sommerfelt et al., Virol. 176:58-59 (1990); Wilson et al., J. Virol. 63:2374-2378 (1989); Miller et al., J. Virol. 65:2220-2224 (1991); PCT/US94/05700).

For applications where transient expression is preferred, adenovirus-based systems can be used. Adenovirus-based vectors are capable of very high transduction efficiency in many cell types and do not require cell division. High titers and high levels of expression have been obtained with such vectors. The vectors can be produced in large quantities in a relatively simple system. Adeno-associated virus ("AAV") vectors are also used to transduce cells with target nucleic acids, e.g., in in vitro production of nucleic acids and peptides, and for in vivo and ex vivo gene therapy procedures (see, e.g., West et al., Virology 160:38-47 (1987); U.S. Pat. No. 4,797,368; WO 93/24641; Kotin, Human Gene Therapy 5:793-801 (1994); Muzyczka, J. Clin. Invest. 94:1351 (1994). Construction of recombinant AAV vectors is described in U.S. Pat. No. 5,173,414; Tratschin et al., Mol. Cell. Biol. 5:3251-3260 (1985); Tratschin et al., Mol. Cell. Biol. 4:2072-2081 (1984); Hermonat and Muzyczka, PNAS USA 81:6466-6470 (1984); and Samulski et al., J. Virol. 63:03822-3828 (1989).

At least six viral vector approaches are available today for gene transfer in clinical trials, which utilize approaches that involve complementation of a defective vector with a gene inserted into a helper cell line to generate the transducing agent.

pLASN and MFG-S are examples of retroviral vectors that have been used in clinical trials (Dunbar et al., Blood 85:3048-305 (1995); Kohn et al., Nat. Med. 1:1017-102 (1995); Malech et al., PNAS USA 94:22 12133-12138 (1997)). PA317/pLASN was the first therapeutic vector used in a gene therapy trial (Blaese et al., Science 270:475-480 (1995)). Transduction efficiencies of 50% or higher have been observed with MFG-S packaged vectors. (Ellem et al., Immunol Immunother. 44(1):10-20 (1997); Dranoff et al., Hum. Gene Ther. 1:111-2 (1997).

Recombinant adeno-associated virus vectors (rAAV) are a promising alternative gene delivery system based on the defective nonpathogenic parvovirus adeno-associated type 2 virus. All vectors are derived from plasmids that carry only the AAV 145 bp inverted terminal repeats flanking the transgene expression cassette. Efficient gene transfer and stable transgene delivery due to integration into the genome of transduced cells are key features of this vector system. (Wagner et al., Lancet 351:9117 1702-3 (1998); Kearns et al., Gene Ther. 9:748-55 (1996)). Other AAV serotypes may also be used in accordance with the present invention, including AAV1, AAV3, AAV4, AAV5, AAV6, AAV8, AAV8.2, AAV9, AAVrhlO, and pseudotyped AAVs such as AAV2/8, AAV2/5, and AAV2/6.

Replication-deficient recombinant adenoviral vectors (Ad) can be produced at high titers and can easily infect several different cell types. Most adenoviral vectors are engineered so that a transgene replaces the Ad E1a, E1b and/or E3 genes, and the replication-deficient vector is then propagated in human 293 cells supplying the deleted gene function in trans. Ad vectors can transduce multiple types of tissues in vivo, including non-dividing, differentiated cells such as those found in liver, kidney and muscle. Conventional Ad vectors have a large carrying capacity. An example of the use of Ad vectors in clinical trials included polynucleotide therapy for antitumor immunization by intramuscular injection (Sterman et al., Hum. Gene Ther. 7:1083-9 (1998)). Further examples of the use of adenoviral vectors for gene transfer in clinical trials include Rosenecker et al., Infection 24:15-10 (1996); Sterman et al., Hum. Gene Ther. 9:7, 1083-1089 (1998); Welsh et al., Hum. Gene Ther. 2:205-18 (1995); Alvarez et al., Hum. Gene Ther. 5:597-613 (1997); Topf et al., Gene Ther. 5:507-513 (1998); Sterman et al., Hum. Gene Ther. 7:1083-1089 (1998).

Packaging cells are used to form viral particles capable of infecting host cells. Such cells include 293 cells, which package adenovirus and AAV, and ψ2 or PA317 cells, which package retrovirus. Viral vectors used in gene therapy are usually generated by producer cell lines that package nucleic acid vectors into viral particles. The vectors generally contain the minimum viral sequences required for packaging and subsequent integration into the host (if applicable), with other viral sequences being replaced by expression cassettes that code for proteins to be expressed. Missing viral functions are supplied in trans by the packaging cell line. For example, AAV vectors used in gene therapy generally only possess the inverted terminal repeat (ITR) sequences from the AAV genome that are necessary for packaging and integration into the host genome. The viral DNA is packaged into a cell line that contains a helper plasmid that codes for the other AAV genes, namely rep and cap, but lacks the ITR sequences. The cell line is also infected with adenovirus as a helper. The helper virus promotes the replication of the AAV vector and the expression of AAV genes from the helper plasmid. The helper plasmid is not packaged in significant amounts due to the lack of ITR sequences. Contamination with adenovirus can be reduced, for example, by heat treatment, to which adenovirus is more sensitive than AAV. Additionally, AAV can be produced using the baculovirus system (see, for example, U.S. Patent Nos. 6,723,551 and 7,271,002).

Purification of AAV particles from the .293 or baculovirus systems typically involves growing cells that produce the virus, followed by recovery of the virus particles from the cell supernatant or lysing the cells and recovering the virus from the crude lysate. The AAV is then purified by methods known in the art, including ion exchange chromatography (see, e.g., U.S. Pat. Nos. 7,419,817 and 6,989,264), ion exchange chromatography and purification using CsCl density centrifugation (e.g., PCT Publication WO2011094198A10), immunoaffinity chromatography (e.g., WO2016128408), or AVB Sepharose (e.g., GE Healthcare Life Sciences).

In many gene therapy applications, it is desirable for the gene therapy vector to be delivered with a high degree of specificity to a particular tissue type. Thus, viral vectors can be modified to have specificity for a given cell type by expressing a ligand as a fusion protein with a viral coat protein on the outer surface of the virus. The ligand is selected to have affinity for a receptor known to be present in the cell type of interest. For example, Han et al. (Proc. Natl. Acad. Sci. USA 92:9747-9751 (1995)) reported that Moloney murine leukemia virus can be modified to express human heregulin fused to gp70, and that the recombinant virus infects certain human breast cancer cells expressing the human epidermal growth factor receptor. This principle can be extended to other virus-target cell pairs where the target cell expresses a receptor and the virus expresses a fusion protein containing a ligand for the cell surface receptor. For example, filamentous phage can be engineered to display antibody fragments (e.g., FAB or Fv) with specific binding affinity to virtually any cellular receptor of choice. Although the above description applies primarily to viral vectors, the same principles can be applied to non-viral vectors. Such vectors can be engineered to contain specific uptake sequences that favor uptake by specific target cells.

Gene therapy vectors can be delivered in vivo by administration to an individual patient, typically by systemic administration (e.g., intravenous, intraperitoneal, intramuscular, subcutaneous or intracranial injection) or local application, as described below. Alternatively, vectors can be delivered ex vivo to cells, e.g., cells explanted from an individual patient (e.g., lymphocytes, bone marrow aspirates, tissue biopsies), or to universal donor hematopoietic stem cells, which can then be reimplanted into the patient, typically after selection of cells that have taken up the vector.

Ex vivo cell transfection (e.g., via re-infusion of transfected cells into a host organism) for diagnostics, research, transplantation, or for gene therapy is well known to those of skill in the art. In a preferred embodiment, cells are isolated from a subject organism, transfected with DNA-binding proteins nucleic acid (gene or cDNA), and re-infused back into the subject organism (e.g., a patient). A variety of cell types suitable for ex vivo transfection are well known to those of skill in the art (see, e.g., Freshney et al., Culture of Animal Cells, A Manual of Basic Technique (3rd ed., 1994) and references cited therein for a discussion of how to isolate and culture cells from a patient).

In one embodiment, stem cells are used in ex vivo procedures for cell transfection and gene therapy. The advantage of using stem cells is that they can be differentiated into other cell types in vitro or can be introduced into a mammal (such as a cell donor) where they engraft in the bone marrow. Methods are known for differentiating CD34+ cells in vitro into clinically important immune cell types using cytokines such as GM-CSF, IFN-γ, and TNF-α (see Inaba et al., J. Exp. Med. 176:1693-1702 (1992)).

Stem cells are isolated using known methods for transduction and differentiation. For example, stem cells are isolated from bone marrow cells by panning the bone marrow cells with antibodies that bind to unwanted cells such as CD4+ and CD8+ (T cells), CD45+ (pan B cells), GR-1 (granulocytes), and Iad (differentiated antigen presenting cells) (see Inaba et al., J. Exp. Med. 176:1693-1702 (1992)).

Modified stem cells can also be used in some embodiments. For example, neuronal stem cells that have been made resistant to apoptosis can be used as therapeutic compositions if the stem cells also contain a ZFP TF of the invention. Resistance to apoptosis can be created in stem cells by knocking out BAX and/or BAK, for example, using BAX-specific ZFNs or BAK-specific ZFNs (see U.S. Patent Application No. 12/456,043), or those that are disrupted in caspases, for example, again using caspase 6-specific ZFNs. These cells can be transfected with ZFP TFs known to regulate the TCR.

Vectors (e.g., retroviruses, adenoviruses, liposomes, etc.) containing therapeutic DNA-binding proteins (or nucleic acids encoding these proteins) can also be administered directly to the organism for transduction of cells in vivo. Alternatively, naked DNA can be administered. Administration is by any of the routes commonly used to introduce molecules into ultimate contact with blood or tissue cells, including but not limited to injection, infusion, topical application, and electroporation. Suitable methods of administering such nucleic acids are available and well known to those of skill in the art, and while more than one route can be used to administer a particular composition, certain routes can often provide a more immediate and effective response than another route.

Methods for the introduction of DNA into hematopoietic stem cells are disclosed, for example, in U.S. Patent No. 5,928,638. Vectors useful for the introduction of transgenes into hematopoietic stem cells, e.g., CD34+ cells, include adenovirus type 35.

Suitable vectors for the introduction of transgenes into immune cells (e.g., T cells) include non-integrating lentiviral vectors. See, e.g., Ory et al. (1996) Proc. Natl. Acad. Sci. USA 93:11382-11388; Dull et al. (1998) J. Virol. 72:8463-8471; Zuffery et al. (1998) J. Virol. 72:9873-9880; Follenzi et al. (2000) Nature Genetics 25:217-222.

Pharmaceutically acceptable carriers are determined, in part, by the particular composition being administered, as well as by the particular method used to administer the composition. Thus, as described below, a variety of suitable formulations of pharmaceutical compositions are available (see, e.g., Remington's Pharmaceutical Sciences, 17th ed., 1989).

As noted above, the disclosed methods and compositions can be used in any type of cell, including, but not limited to, prokaryotic cells, fungal cells, archaeal cells, plant cells, insect cells, animal cells, vertebrate cells, mammalian cells, and human cells, including T cells and stem cells of any type. Suitable cell lines for protein expression are known to those of skill in the art and include, but are not limited to, insect cells such as COS, CHO (e.g., CHO-S, CHO-K1, CHO-DG44, CHO-DUXB11), VERO, MDCK, WI38, V79, B14AF28-G3, BHK, HaK, NS0, SP2/0-Ag14, HeLa, HEK293 (e.g., HEK293-F, HEK293-H, HEK293-T), perC6, Spodoptera fugiperda (Sf), and fungal cells such as Saccharomyces, Pichia, and Schizosaccharomyces. Progeny, variants, and derivatives of these cell lines can also be used.

Applications The disclosed compositions and methods can be used for any application in which it is desirable to modulate B2M expression and/or functionality, including, but not limited to, therapeutic and research applications in which B2M modulation is desirable. For example, the disclosed compositions can be used in vivo and/or ex vivo (cell therapy) to disrupt expression of endogenous B2M in T cells engineered to express one or more exogenous CAR, TCR, ACTR or other cancer-specific receptor molecules for adoptive cell therapy, thereby treating and/or preventing cancer. Furthermore, in such situations, modulation of B2M expression in cells can eliminate or substantially reduce the risk of unwanted cross-reaction with healthy, non-targeted tissues (i.e., graft-versus-host response).

The methods and compositions also include stem cell compositions, where the B2M gene in the stem cell is modulated (altered) and the cells further comprise ACTR and/or CAR and/or isolated or engineered TCR. For example, B2M knockout or knockdown modulated allogeneic hematopoietic stem cells can be introduced into HLA-matched patients after bone marrow ablation. These altered HSCs allow for repopulation in the patient but do not result in potential GvHD. The introduced cells can also have other modifications (e.g., chemotherapy resistance) that are useful during subsequent therapy to treat the underlying disease. HLA-null cells also have application as an "off the shelf" therapy in emergency room settings for trauma patients.

The methods and compositions of the invention are also useful for designing and implementing in vitro and in vivo models, e.g., animal models, of B2M and/or HLA and related disorders, enabling the study of these disorders.

All patents, patent applications and publications referred to herein are hereby incorporated by reference in their entireties.

Although the disclosure has been provided in some detail by way of illustration and example for clarity of understanding, it is apparent to one skilled in the art that various changes and modifications can be made without departing from the spirit or scope of the disclosure. Therefore, the foregoing disclosure and the following examples should not be construed as limiting.

Example 1
Design of B2M-specific nucleases B2M-specific ZFNs were constructed to allow for site-specific introduction of double-stranded breaks in the B2M gene. ZFNs were designed essentially as described in Urnov et al. (2005) Nature 435(7042):646-651, Lombardo et al. (2007) Nat Biotechnol. Nov;25(11):1298-306 and US Patent Publication Nos. 2008-0131962, 2015-016495, 2014-0120622, 2014-0301990 and US Patent No. 8,956,828. ZFN pairs targeted different sites in the constant region of the B2M gene (see FIG. 1). The recognition helices as well as target sequences for exemplary ZFN pairs are shown in Table 1 below. The target site for the B2M zinc finger design is shown in the first column. Nucleotides in the target site targeted by the ZFP recognition helix are shown in uppercase; non-targeted nucleotides are shown in lowercase. The linker used to join the FokI nuclease domain and the ZFP DNA binding domain is also shown (see US Patent Publication No. 20150132269). For example, the amino acid sequence of domain linker L0 is: DNA binding domain-QLVKS-FokI nuclease domain (SEQ ID NO:3). Similarly, the amino acid sequence for domain linker N7a is: FokI nuclease domain-SGTPHEVGVYTL-DNA binding domain (SEQ ID NO:4) and N6a is: FokI nuclease domain-SGAQGSTLDF-DNA binding domain (SEQ ID NO:5).

All ZFNs were tested and found to bind to their target sites and to be active as nucleases.

Guide RNAs for the CRISPR/Cas9 system of S. pyogenes were also constructed to target the B2M gene. The target sequence in the B2M gene as well as the guide RNA sequence are shown in Table 2A below. All guide RNAs are tested in the CRISPR/Cas9 system and found to be active. The lowercase "g" at the 5' end of some guide sequences indicates an added G nucleotide that plays a role in the PAM sequence.

TALENs were generated to target the B2M locus and are shown in Table 2B below. All TALENs were tested in K562 cells and found to be active (see Table 2C and FIG. 2B).

The TALENs from Table 2B were tested at three different concentrations, either 25, 100 or 400 ng of each TALEN per reaction. All TALENs tested were found to bind to their target sites and were found to be active as nucleases; exemplary data are shown in Table 2C and Figure 2B.

Thus, the nucleases described herein (e.g., nucleases comprising a ZFP, TALE, or sgRNA DNA binding domain) bind to their target sites and cleave the B2M gene, thereby creating genetic modifications in the B2M gene, including any of SEQ ID NOs: 6-48 or 137-205, including modifications (insertions and/or deletions) within any of these sequences (SEQ ID NOs: 6-48 or 137-205); modifications within 1-50 (e.g., 1 to 10) base pairs of these gene sequences; modifications between target sites of paired target sites (for dimers); and/or modifications within one or more of the following sequences: GGCCTTA, TCAAATT, TCAAAT, TTACTGA, and/or AATTGAA (see FIG. 1).

Furthermore, the DNA binding domains (ZFPs, TALEs and sgRNAs) also all bind to their target sites and assemble into active engineered transcription factors when associated with one or more transcriptional regulatory domains.

Example 2
B2M-specific ZFN activity in T cells B2M-specific ZFN pairs were tested in human T cells for nuclease activity. mRNA encoding the ZFNs was transfected into purified T cells. Briefly, T cells were obtained from leukopheresis products and purified using the Miltenyi CliniMACS system (CD4 and CD8 double selection). These cells were then activated using Dynabeads (ThermoFisher) according to the manufacturer's protocol. Three days after activation, cells were transfected with two doses of mRNA (2 or 6 μg total of two ZFNs) using a BTX 96-well electroporator (BTX) according to standard protocols. Cells were then expanded for an additional 7 days, for a total of 10 days after activation. Cells were harvested on day 7 and analyzed for targeted B2M alterations using deep sequencing (Miseq, Illumina) and on day 10 by FACs analysis using HLA-A, -B and -C staining.

B2M-specific ZFN pairs were all active in T cells, yielding an average of 89% and 83% for 6 μg and 2 μg mRNA doses, respectively (see FIG. 2). The pairs and positions used (shown in FIG. 1) are listed in Table 3 below.

Similarly, T cells treated with ZFNs lost expression of HLA A, B and C, where FACS analysis showed an average of 81% and 67% HLA-negative T cells at 6 μg and 2 μg mRNA doses, respectively (see Figure 3).

Example 3
In vitro activity of guide RNA against B2M In these experiments, Cas9 was provided in a pVAX plasmid and sgRNA was provided in a plasmid under the control of the U6 promoter. Plasmids were mixed at either 100ng each or 400ng each and mixed with 2e5 cells per run. Cells were transfected using the Amaxa system. Briefly, nucleic acids were transfected using the Amaxa transfection kit and the standard Amaxa shuttle protocol. After transfection, cells were rested for 10 minutes at room temperature and then resuspended in pre-warmed RPMI. Cells were then grown in standard conditions at 37°C. Genomic DNA was isolated 7 days after transfection and subjected to MiSeq analysis.

The data presented below (Table 4) shows the percentage of indels (insertions and deletions) detected at two doses of guide RNA, indicating that the various guide RNAs induce cleavage at the targeted site. Numbers represent the average of two experiments. All guides were active.

Example 4
B2M and TCR double knockout in primary T cells The B2M pair described herein was also tested in combination with ZFNs specific for TCRA (see Table 5 below and US Provisional Patent Applications 62/269,365 and 62/306,500). Cells were obtained and treated as described in Example 2. The mRNAs encoding the ZFN pairs (SBS#57017/SBS#57327 for B2M and SBS#55254/SBS#55248 for TCRA) were electroporated into cells using a Maxcyte instrument according to the manufacturer's instructions. Briefly, T cells were activated on day 0 and treated with ZFN-encoding mRNA on day 3 at a cell density of 3e7 cells/mL. Electroporation was followed by a 30° C. cold shock overnight after electroporation. On day 4, cells were counted, assayed for viability, diluted to 0.5e6 cells/mL, and transferred to 37° C. On day 7, cells were counted, assayed again, and rediluted to 0.5e6 cells/mL. On days 10 and 14, a portion of cells was harvested for FACS and MiSeq deep sequencing.

Cells were divided into four groups: control without ZFN, TCRA ZFN only, B2M ZFN only, and TCRA ZFN + B2M ZFN. FACS analysis showed that the sets of ZFNs against TCRA alone (180 μg/μL ZFN mRNA) or B2M alone (180 μg/μL ZFN mRNA) resulted in a high rate of cleavage (96% CD3 marking for TCRA ZFN and 92% knockout of HLA marking for B2M ZFN (Figure 4)). When cells were treated with both types of ZFN pairs (both at 180 μg/μL), 82% of cells lost both CD3 and HLA marking.

Similar groups of cells were treated with various amounts of TCRA-specific ZFNs as indicated (60-250ug/uL) plus 60ug/mL of B2M ZFN and subjected to MiSeq deep sequencing (Illumina) and FACs analysis on days 10 and 14, with the results shown in Table 6 below. The results show that a high rate of double knockouts, as detected by NHEJ-mediated insertions and deletions, was observed with these ZFNs.

Thus, this data demonstrates that a double knockout of B2M and TCRA at or near the target and/or within 1-50 (e.g., 1-10) base pairs (including between the paired target sites) of the target sequence and/or cleavage site of the nucleases described herein (including the B2M sequence TCAAAT (site D in Figure 1) and the TCRA sequence CCTTC between the two target sequences for the SBS#55254/SBS#55248 TCRA-specific pair) inactivated both genes.

Example 5
Double knockout of B2M and TCRA by targeted integration In this experiment the TCRA-specific ZFN pair was SBS No. 55266/SBS No. 53853, which contains the sequence TTGAAA between the TCRA-specific ZFN target sites (Table 5), and the B2M pair was SBS No. 57332/SBS No. 57327, which contains the sequence TCAAAT between the B2M-specific ZFN target sites (Table 1).

Briefly, T cells (AC-TC-006) were thawed and activated with CD3/28 dynabeads (1:3, cell:bead ratio) in X-vivo15 T cell culture medium (day 0). After two days of culture (day 2), AAV donors (containing a GFP transgene and homology arms to TCRA or B2M genes) were added to the cell cultures, except that a donor-free control group was also maintained. The next day (day 3), TCRA and B2M ZFNs were added in the following five groups via mRNA delivery:
(a) Group 1 (TCRA and B2M ZFNs only, no donor): TCRA 120ug/mL: B2M only 60ug/mL;
(b) Group 2 (donor with TCRA and B2M ZFNs and TCRA homology arms): TCRA 120ug/mL; B2M 60ug/mL and AAV(TCRA-hPGK-eGFP-Clone E2) 1E5vg/cell;
(c) Group 3 (donor with TCRA and B2M ZFNs and TCRA homology arms): TCRA 120ug/mL; B2M 60ug/mL; and AAV(TCRA-hPGK-eGFP-Clone E2)3E4vg/cell;
(d) Group 4 (donor with TCRA and B2M ZFNs and B2M homology arms): TCRA 120ug/mL; B2M 60ug/mL and AAV (pAAV B2M-site D-hPGK GFP) 1E5vg/cell (e) Group 5 (donor with TCRA and B2M ZFNs and B2M homology arms): TCRA 120ug/mL; B2M 60ug/mL and AAV (pAAV B2M-site D-hPGK GFP) 3E4vg/cell.
All experiments were performed using the protocol described in US Patent Application No. 15/347,182 (extreme cold shock) at a cell density of 3e7 cells/ml and incubated for cold shock at 30° C. overnight after electroporation.

The next day (day 4), cells were diluted to 0.5e6 cells/ml and transferred to culture at 37°C. Three days later (day 7), cells were diluted again to 0.5e6 cells/ml. After an additional 3 and 7 days of culture (days 10 and 14, respectively), cells were harvested (diluted to 0.5e6 cells/ml) for FACS and MiSeq analysis.

As shown in FIG. 5, GFP expression indicated successful targeted integration and resulted in genetically modified cells containing B2M and TCRA modifications (insertions and/or deletions) within the nuclease site (or within 1 to 50 base pairs of the nuclease target site, e.g., TTGAAA and TCAAAT, and/or between the paired target sites).

Experiments will also be performed in which the CAR transgene is integrated into the B2M and/or TCRA locus to generate a double B2M/TCRA knockout expressing CAR.

All patents, patent applications and publications mentioned herein are hereby incorporated by reference in their entirety.

Although the disclosure has been provided in some detail by way of illustration and example for clarity of understanding, it is apparent to one skilled in the art that various changes and modifications can be made without departing from the spirit or scope of the disclosure. Therefore, the above description and examples should not be construed as limiting.

Claims (13)

A fusion molecule comprising a DNA binding domain and a CRISPR/Cas system,
A fusion molecule comprising a single guide RNA (sgRNA) wherein the DNA binding domain binds to exon 1 or exon 2 of the B2M gene, and the DNA binding domain comprises a nucleotide sequence selected from any one of SEQ ID NOs: 105-107 or 110-111 . A polynucleotide encoding the fusion molecule of claim 1. The polynucleotide of claim 2, which is a viral vector, a plasmid, or an mRNA. An isolated cell comprising the fusion molecule of claim 1. An isolated cell comprising the polynucleotide of claim 2. An isolated cell derived from the cell of claim 4, wherein expression of the beta 2 microglobulin (B2M) gene is partially or completely inactivated in the cell by insertions and/or deletions within GGCCTTA, TCAAAT, TCAAATT, TTACTGA and/or AATTGAA in exon 1 or exon 2 of the B2M gene. The isolated cell of claim 6, further comprising an inactivated T cell receptor gene, a PD1 gene and/or a CTLA4 gene. The isolated cell of claim 6, further comprising a transgene encoding a chimeric antigen receptor (CAR), an antibody-binding T cell receptor (ACTR), and/or an engineered TCR. The isolated cell of claim 6, wherein the cell is a lymphocytic cell, a stem cell, or a progenitor cell. The cell according to claim 9, wherein the cell is a T cell, an induced pluripotent stem cell (iPSC), a mesenchymal stem cell (MSC), or a hematopoietic stem cell (HSC). An isolated population of cells comprising the isolated cell of claim 6. A pharmaceutical composition comprising an isolated population of cells according to claim 11. The pharmaceutical composition according to claim 12 for treating or preventing cancer in a subject having cancer.

Images